A Historical Perspective on India’s Inflation Persistence: A Quantile Analysis

Taniya Ghosh; Yadavindu Ajit

doi:10.1515/snde-2024-0078

Article

A Historical Perspective on India’s Inflation Persistence: A Quantile Analysis

Taniya Ghosh and Yadavindu Ajit

Published/Copyright: August 25, 2025

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information

From the journal Studies in Nonlinear Dynamics & Econometrics

Abstract

This study investigates historical inflation persistence in India under three distinct regimes: monetary targeting, multiple indicators, and inflation targeting (IT). Previous studies on India often relied on mean-based techniques, which can be biased when inflation exhibits a skewed distribution. To address this, we employ a quantile-based approach to examine inflation persistence. Our findings reveal asymmetric inflation behavior, with varying persistence depending on the magnitude of the inflation-affecting shock. Inflation persistence was notably high during the multiple indicators regime but declined following the adoption of IT, especially in the pre-COVID period. While IT helped reduce persistence, it was less effective during COVID-19. Using a wavelet-based method, we also find that inflation persistence in India is closely linked to the persistence of food inflation. Given that food price shocks exhibit long-lasting effects, our findings support the case for targeting headline inflation, contributing to the policy debate in India on whether to target headline or core inflation. Overall, our results highlight the evolving nature of inflation persistence across regimes and emphasize the need for complementary supply-side measures to enhance the efficacy of monetary policy.

Keywords: inflation persistence; monetary regime; quantile regression

JEL Classification: C21; E31; E52

Corresponding author: Taniya Ghosh, Indira Gandhi Institute of Development Research (IGIDR), Gen. A. K. Vaidya Marg, Filmcity Road, Mumbai, 400065, India, E-mail: taniya@igidr.ac.in

Appendix A: Inflation and its Measurement in India

In India, inflation is measured primarily using the Wholesale Price Index (WPI) and the Consumer Price Index for Industrial Workers (CPI-IW), along with their disaggregates. Before the introduction of new CPI measures, WPI and CPI-IW were the key indicators in academic and policy discussions.^[26] The WPI and CPI-IW differ significantly in their weighting. CPI-IW assigns 40 % to food, making it highly sensitive to food price fluctuations, while WPI gives only 16 % to food but a higher weight (14.23 %) to fuel, making it more sensitive to global oil prices. Additionally, CPI-IW includes services, whereas WPI does not, contributing to inflation rate differences. These variations highlight the distinct sensitivities of each index, underscoring the importance of careful interpretation when comparing inflation rates.

Table A1:

Weights of different products under different base years for WPI.

#	Major groups/groups	1970–71	1981–82	1993–94	2004–05	2011–12
	All commodities	100.00	100.00	100.00	100.00	100.00
1	Primary articles	41.67	32.30	22.03	20.12	22.62
2	Fuel and power	8.46	10.66	14.23	14.91	13.15
3	Manufactured products	49.87	57.04	63.75	64.97	64.23

Source: Office of the Economic Advisor, Ministry of Commerce.

Table A2:

CPI-IW disaggregates weights.

Groups	Labour bureau
	1982	2001	2016
I-A. Food & beverages (food group)^*	57.00	46.20	39.17
I-B. Pan, supari, tobacco & intoxicants	3.15	2.27	2.07
II. Fuel & light	6.28	6.43	5.50
III. Housing	8.67	15.27	16.87
IV. Clothing & footwear (clothing, bedding & footwear)^*	8.54	6.57	6.08
V. Miscellaneous	16.36	23.26	30.31

Notes: *The bracketed text shows the name of the category in the previous CPI-IW series.

Appendix B: Rolling Regression

For our preliminary analysis, we use rolling regression considering a 5-year rolling period while using the bootstrap method to generate a confidence band. Our rolling regression analysis estimates an AR(p) model, as shown below.

Let π _t measure inflation, α an intercept term, and ϵ _t be a serially uncorrelated error term. Consider an AR(p) process given by:

(7) π t = α + ∑ k = 1 p β k π t − k + ϵ t

where inflation persistence (ρ) = ∑ k = 1 p β k . Equation (7) can be re-written in difference terms as follows:

(8) π t = α + ρ π t − 1 + ∑ k = 1 p − 1 γ k Δ π t − k + ϵ t

where Δπ _t = π _t − π _t−1. The parameter ρ in Equation (8) holds significant implications for the nature of the inflation process.^[27] When the absolute value of ρ, denoted as |ρ|, equals 1 (i.e. |ρ| = 1), it indicates that the inflation rate possesses a unit root, resulting in a random walk process. This implies infinite persistence, as shocks to the inflation rate have a lasting impact and do not dissipate over time. Conversely, if |ρ| is less than 1 (i.e. |ρ| < 1), the inflation rate exhibits mean-reverting characteristics following a shock, indicating a stationary process. In a stationary process, shocks to the inflation rate eventually fade away, and the inflation rate returns to its long-term equilibrium level. The magnitude of ρ thus provides valuable information about the persistence and stationarity of the inflation process. The estimates of ρ in Equation (8) can be obtained from least square estimations. However, the least square estimation suffers from a bias as ρ approaches unity; therefore, we also estimate the confidence band of ρ following Hansen (1999).

Appendix C: RWWQC Methodology

To investigate the dynamic dependence of inflation disaggregates persistence on inflation persistence across different time horizons and quantiles, following Özkan et al. (2024), we employ the Rolling Window Wavelet Quantile Correlation (RWWQC). This methodology extends the Wavelet Correlation (WC) proposed by Gençay et al. (2002) and the Rolling Window Wavelet Correlation (RWWC) introduced by Polanco-Martínez et al. (2018), by incorporating quantile-specific dependence structures to capture heterogeneous co-movements across the distribution, wherein wavelet quantile correlation (WQC) suggested by Kumar and Padakandla (2022) is used within a rolling window framework.

C.1 Wavelet Correlation (WC)

The wavelet correlation between two time series X _t and Y _t at scale j captures their dependence at specific frequency ranges. Following Gençay et al. (2002), the wavelet correlation at scale s _j is defined as:

(C.1) ρ X Y ( s j ) = δ X Y ( s j ) σ X 2 ( s j ) σ Y 2 ( s j ) ,

where δ _XY(s _j) represents the wavelet covariance between X _t and Y _t, while σ X 2 ( s j ) and σ Y 2 ( s j ) are the corresponding wavelet variances.

C.2 Rolling Window Wavelet Correlation (RWWC)

To capture temporal changes in the wavelet correlation, Polanco-Martínez et al. (2018) introduced the Rolling Window Wavelet Correlation (RWWC). This approach estimates WC within a moving window of size w, shifting forward one observation at a time. The rolling wavelet correlation at scale j within window w _i is given by:

(C.2) ρ ̃ X Y ( λ j , w i ) = γ ̃ X Y ( λ j , w i ) σ ̃ X ( λ j , w i ) σ ̃ Y ( λ j , w i ) ,

where γ ̃ X Y ( λ j , w i ) is the unbiased wavelet covariance estimator in window w _i, and σ ̃ X 2 ( λ j , w i ) , σ ̃ Y 2 ( λ j , w i ) are the wavelet variances.

C.3 WQC

RWWC is further combined with WQC. Li et al. (2015) proposed quantile correlation; further, Kumar and Padakandla (2022) extended and proposed WQC. The quantile covariance (QC) between X _t and Y _t, at different quantiles, is as follows. Let Q _τ,X represent the τth quantile of X, and let Q _τ,Y(X) denote the τth quantile of Y conditional on X. The conditional quantile function, Q _τ,Y(X), is independent of X if and only if the random variables I(Y − Q _τ,Y > 0) and X are independent (Koenker 2005). Here, I(.) is an indicator function that takes the value 1 if its argument is true and 0 otherwise. For 0 < τ < 1, the quantile covariance function is defined as:

(C.3) q δ τ ( Y , X ) = cov I ( Y − Q τ , Y > 0 ) , X = E ϕ τ ( Y − Q τ , Y ) ( X − E [ X ] ) ,

where the function ϕ _τ(w) is given by:

(C.4) ϕ τ ( w ) = τ − I ( w < 0 ) .

C.3.1 Quantile Covariance

Following Li et al. (2015), the quantile covariance (QC) between two time series X _t and Y _t at the conditional quantile level τ is given by:

(C.5) QC τ ( X , Y ) = q δ τ ( Y , X ) σ 2 ( ϕ τ ( Y − Q τ , Y ) ) σ 2 ( X ) , 0 < τ < 1 ,

where:

qδ _τ(Y, X) represents the quantile-specific covariance.
Q _τ,Y is the conditional quantile of Y at level τ.
ϕ _τ is a quantile transformation function capturing asymmetric dependence.
σ ²(ϕ _τ(Y − Q _τ,Y)) and σ ²(X) denote the variances of transformed Y and X, respectively.

This measure allows us to analyze dependencies not just at the mean but across different quantiles, providing a more comprehensive view of co-movement patterns in extreme conditions such as financial crises or periods of rapid inflation.

C.3.2 Wavelet Quantile Covariance (WQC)

To extend quantile covariance analysis to different time horizons, Kumar and Padakandla (2022) introduced the Wavelet Quantile Covariance (WQC). This approach applies the Maximal Overlap Discrete Wavelet Transform (MODWT) to decompose the time series into multiple frequency components before computing quantile covariance at each scale.^[28] In this study, we employ Multiresolution Analysis (MRA) using MODWT to decompose the time series data into multiple frequency components. The decomposition is performed using the least asymmetric Daubechies wavelet filter of length L = 8, denoted LA(8) across five resolution levels (J = 5).^[29] The MODWT is preferred over the standard Discrete Wavelet Transform (DWT) as it preserves time alignment by avoiding downsampling, making it particularly suitable for time series analysis. The MRA decomposition yields a set of wavelet coefficients corresponding to different frequency bands, allowing us to isolate high-frequency noise from lower-frequency trends. This methodology enables a nuanced understanding of the underlying temporal patterns in the data while preserving its structural integrity across different time scales.

The WQC between two time series X _t and Y _t at scale j and conditional quantile τ is given by:

(C.6) WQC τ ( s j [ X ] , s j [ Y ] ) = q δ τ ( s j [ Y ] , s j [ X ] ) σ 2 ( ϕ τ ( s j [ Y ] − Q τ , s j [ Y ] ) ) σ 2 ( s j [ X ] ) , 0 < τ < 1 ,

where:

qδ _τ(s _j[Y], s _j[X]) is the quantile-specific covariance at scale s _j.
Q τ , s j [ Y ] represents the conditional quantile of Y at scale s _j.
σ ²(s _j[X]) and σ 2 ( ϕ τ ( s j [ Y ] − Q τ , s j [ Y ] ) ) denote the wavelet-based variances.

Multiple existing studies (Abakah et al. 2024; Jalal and Gopinathan 2023), have used WQC for measuring the relationship between two time series at different frequency levels using wavelet transforms.

C.3.2.1 Wavelet Correlation at Scale s _j

It is defined as:

(C.7) p X Y = δ ̃ X Y ( s j ) σ ̃ X 2 ( s j ) σ ̃ Y 2 ( s j )

where:

δ ̃ X Y ( s j ) is the unbiased estimator of the wavelet covariance at scale s _j,
σ ̃ X 2 ( s j ) and σ ̃ Y 2 ( s j ) are the wavelet variances of X and Y, respectively.

To capture time-varying correlations, rolling window wavelet correlation (RWWC) applies wavelet correlation within a rolling window, allowing for the analysis of dynamic relationships over time (Özkan et al. 2024).

C.3.3 Rolling Window Wavelet Quantile Correlation (RWWQC)

While RWWC captures time-varying dependencies, it assumes a constant correlation across all quantiles. However, economic and financial time series often exhibit asymmetric relationships, where dependencies may differ in the tails of the distribution. To account for this, we extend RWWC to Rolling Window Wavelet Quantile Correlation (RWWQC), allowing for quantile-specific dependence structures.

For a given scale j, rolling window w _i, and quantile level τ, the RWWQC is defined as:

(C.8) RWWQC τ ( s j , w i [ X ] , s j , w i [ Y ] ) = q δ τ ( s j , w i [ Y ] , s j , w i [ X ] ) σ 2 ( ϕ τ ( s j , w i [ Y ] − Q τ , s j , w i [ Y ] ) ) σ 2 ( s j , w i [ X ] ) ,

where:

qδ _τ(s _j, w _i[Y], s _j, w _i[X]) is the quantile covariance at scale s _j and window w _i.
Q τ , s j , w i [ Y ] represents the conditional quantile of Y at level τ.
ϕ _τ is the quantile-specific transformation function capturing asymmetric dependence.
σ ²(s _j, w _i[X]) and σ 2 ( ϕ τ ( s j , w i [ Y ] − Q τ , s j , w i [ Y ] ) ) represent the wavelet variances at scale s _j within window w _i.

This formulation enables us to assess how dependencies evolve not only across time and frequency but also across different quantiles, allowing for a richer characterization of co-movement patterns in financial and economic variables.

Appendix D: Results Tables and Figures

D.1 Regimewise Comparison Analysis for Rolling Regression

Regime wise comparison results as shown in Table D.1 for WPII and Table D.2 for CPII-IW shows negative and significant coefficient for inflation targeting regime suggesting that inflation persistence has declined more compared to other two regimes.

Table D.1:

Regimewise comparison for WPII persistence.

Variables	Model-2	Model-2
	WPI persistence	WPI persistence 7 years
IT_dummy	−0.0980^***	−0.0593^***
	(0.0214)	(0.0146)
time_trend	0.0020^***	0.0016^***
	(0.0001)	(0.0001)
Constant	−0.4234^***	−0.1241^***
	(0.0734)	(0.0455)

WPII persistence is the level of persistence calculated using 5-year mean based rolling regression and WPII persistence 7 years is the level of persistence calculated using 7-year mean based rolling regression. Model 2 result consider multiple indicators regime and IT regime. Notes: *, **, and ***denote statistical significance at the 10 %, 5 %, and 1 % levels, respectively.

Table D.2:

Regimewise comparison results for CPII-IW persistence.

Variables	Model 1	Model 2	Model 3	Model 1	Model 2	Model 3
	ρ ₁	ρ ₁	ρ ₁	ρ ₂	ρ ₂	ρ ₂
MI_dummy	−0.0995^***		−0.0794^***	−0.0255		−0.0091
	(0.0241)		(0.0249)	(0.0183)		(0.0192)
time_trend	0.0006^***	0.0006^***	0.0005^***	0.0004^***	0.0004^***	0.0004^***
	(0.0001)	(0.0001)	(0.0001)	(0.0001)	(0.0001)	(0.0001)
IT_dummy		−0.0432^**	−0.1058^***		−0.0595^***	−0.0588^**
		(0.0206)	(0.0358)		(0.0158)	(0.0283)
Constant	0.5631^***	0.4644^***	0.6077^***	0.6559^***	0.6458^***	0.6919^***
	(0.0399)	(0.0762)	(0.0409)	(0.0318)	(0.0622)	(0.0329)

ρ ₁ is the level of persistence calculated using 5-year mean based rolling regression and ρ ₂ is the level of persistence calculated using 7-year mean based rolling regression. Model 1 result consider the monetary targeting regime and multiple indicators regime. Model 2 consider the multiple indicators regime and the IT regime. Model 3 consider all the three regimes. Notes: *, **, and ***denote statistical significance at the 10 %, 5 %, and 1 % levels, respectively.

D.2 Descriptive Statistics for CPII-IW Persistence

Table D.3:

Summary statistics of CPII-IW and its disaggregates persistence.

	Mean	Std. dev.	Skewness	Kurtosis	JB	ADF	PP
CPII-IW persistence	0.82758	0.1193	−1.1694	4.7236	198.30 ^***	−3.003^**	−24.87^**
CPII-IW-cloth persistence	0.81654	0.1461	−1.1847	4.0572	157.93^***	−3.092^**	−20.26^*
CPII-IW-core persistence	0.71417	0.1908	−1.2389	4.9725	235.61^***	−2.381	−22.49^**
CPII-IW-food persistence	0.83479	0.0947	−1.4010	6.2456	431.83^***	−3.888^***	−40.07^***
CPII-IW-fuel persistence	0.64113	0.4682	−2.1481	7.4564	898.76^***	−2.910^**	−24.02^**
CPII-IW-housing persistence	0.75035	0.1930	−2.7650	15.6335	4,458.72^***	−3.984^***	−42.39^***
CPII-IW-tobacco persistence	0.86437	0.0994	−1.5000	4.8746	293.62^***	−2.983^**	−31.75^***
CPII-IW-miscellaneous persistence	0.82712	0.1111	−0.5187	3.0335	25.20^***	−3.167^**	−22.71^**

Notes: JB = Jarque–Bera statistic; ADF = augmented Dickey–Fuller test statistic; PP = Phillips–Perron test statistic. Superscript ***, ** and * represents p < 0.01, p < 0.05 and p < 0.10 respectively.

Table D.4:

BDS test statistics for CPI-IW and its disaggregates persistence.

Series	Embedding dimension (m)
	m = 2	m = 3	m = 4	m = 5	m = 6
CPII-IW persistence	103.2668^***	165.5883^***	284.3856^***	525.6111^***	1,033.1658^***
CPII-IW-cloth persistence	66.7880^***	100.0329^***	157.8137^***	264.5304^***	467.7554^***
CPII-IW-core persistence	136.2689^***	224.9908^***	399.3242^***	763.2094^***	1,549.5824^***
CPII-IW-food persistence	97.8329^***	157.0741^***	270.9989^***	502.4016^***	985.4269^***
CPII-IW-fuel persistence	39.1992^***	47.8687^***	59.9155^***	77.7237^***	104.2817^***
CPII-IW-housing persistence	86.6724^***	124.7692^***	188.8147^***	302.7823^***	509.9729^***
CPII-IW-tobacco persistence	65.8562^***	93.7519^***	139.2174^***	220.2433^***	367.4028^***
CPII-IW-miscellaneous persistence	415.4387^***	738.7390^***	1,445.3985^***	3,072.6460^***	6,980.5841^***

Notes: BDS test statistics for embedding dimension m. Superscript *** denotes significance at the 1 % level.

Table D.5:

Quantilewise results for CPII-IW and CPII-IW-core.

Monetary targeting regime					Multiple indicators regime					IT regime					Pre-COVID IT regime
Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root
CPII-IW
0.1	0.814	−2.609	−2.461	0	0.1	0.856	−3.226	−2.647	0	0.1	0.529	−16.006	−2.345	0	0.1	−3.067	8.340	−2.310	1
0.15	0.837	−2.727	−2.684	0	0.15	0.844	−3.585	−2.704	0	0.15	0.518	−11.257	−2.378	0	0.15	−3.067	9.445	−2.310	1
0.2	0.791	−4.076	−2.751	0	0.2	0.899	−2.435	−2.929	1	0.2	0.469	−7.061	−2.634	0	0.2	−3.067	11.485	−2.310	1
0.25	0.785	−4.914	−2.734	0	0.25	0.902	−2.617	−3.030	1	0.25	0.556	−4.046	−2.310	0	0.25	−3.067	−9.269	−2.310	0
0.3	0.782	−4.298	−2.728	0	0.3	0.912	−2.475	−3.172	1	0.3	0.590	−3.498	−2.679	0	0.3	−3.067	−9.983	−2.310	0
0.35	0.794	−4.161	−2.905	0	0.35	0.929	−2.041	−3.188	1	0.35	0.663	−2.450	−2.933	1	0.35	−3.067	−10.661	−2.310	0
0.4	0.826	−3.717	−3.139	0	0.4	0.961	−1.203	−3.122	1	0.4	0.688	−2.293	−3.210	1	0.4	−3.104	−6.428	−2.310	0
0.45	0.835	−3.700	−3.036	0	0.45	0.970	−0.996	−3.094	1	0.45	0.725	−2.070	−2.925	1	0.45	−4.171	−8.407	−2.573	0
0.5	0.861	−2.760	−3.146	1	0.5	0.976	−0.720	−2.949	1	0.5	0.719	−1.961	−2.978	1	0.5	−4.171	−8.522	−2.310	0
0.55	0.879	−2.327	−3.378	1	0.55	0.985	−0.397	−3.081	1	0.55	0.730	−1.785	−2.827	1	0.55	−4.171	−8.407	−2.310	0
0.6	0.892	−1.866	−3.184	1	0.6	0.973	−0.765	−3.119	1	0.6	0.683	−2.250	−2.870	1	0.6	−4.171	−8.098	−2.870	0
0.65	0.907	−1.529	−3.288	1	0.65	0.961	−0.937	−2.878	1	0.65	0.651	−3.091	−2.933	0	0.65	−4.171	−7.669	−2.310	0
0.7	0.915	−1.249	−3.170	1	0.7	0.963	−0.918	−2.756	1	0.7	0.661	−3.297	−2.668	0	0.7	−4.595	−8.121	−2.310	0
0.75	0.933	−0.931	−3.264	1	0.75	0.958	−1.017	−2.771	1	0.75	0.689	−3.384	−2.881	0	0.75	−4.595	−16.611	−2.310	0
0.8	0.908	−0.998	−3.129	1	0.8	1.023	0.564	−2.853	1	0.8	0.800	−3.108	−2.912	0	0.8	−4.595	64.750	−2.310	1
KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV		KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root
		4.91	2.99	0			3.58	3.00	0			16.01	3.02	0			64.75	3.00	0
CPII-IW-core
0.1	0.404	−8.896	−2.944	0	0.1	0.713	−5.205	−2.340	0	0.1	0.753	−1.456	−3.202	1	0.1	−0.376	23.018	−3.164	1
0.15	0.419	−5.959	−2.667	0	0.15	0.782	−4.712	−2.414	0	0.15	0.774	−1.404	−2.854	1	0.15	−0.376	26.070	−3.171	1
0.2	0.420	−5.517	−2.891	0	0.2	0.849	−3.416	−2.910	0	0.2	0.807	−1.083	−2.800	1	0.2	−0.376	28.740	−3.311	1
0.25	0.485	−4.500	−3.057	0	0.25	0.868	−3.003	−2.899	0	0.25	0.799	−1.163	−2.664	1	0.25	−0.376	−26.845	−2.803	0
0.3	0.508	−4.611	−3.082	0	0.3	0.863	−3.111	−2.988	0	0.3	0.788	−1.261	−2.899	1	0.3	−0.376	−25.616	−2.878	0
0.35	0.522	−4.339	−3.196	0	0.35	0.885	−2.811	−3.050	1	0.35	0.853	−1.726	−2.547	1	0.35	−0.376	−16.827	−2.310	0
0.4	0.600	−3.723	−2.993	0	0.4	0.923	−2.101	−2.915	1	0.4	0.916	−1.143	−2.836	1	0.4	−0.376	−17.769	−2.310	0
0.45	0.563	−3.348	−3.131	0	0.45	0.935	−1.773	−3.010	1	0.45	0.872	−2.052	−2.635	1	0.45	−0.366	−18.310	−2.310	0
0.5	0.535	−3.571	−3.000	0	0.5	0.944	−1.532	−2.998	1	0.5	0.884	−1.804	−2.625	1	0.5	−0.398	−18.996	−2.310	0
0.55	0.561	−3.202	NaN	0	0.55	0.945	−1.472	−3.069	1	0.55	0.880	−1.861	−2.808	1	0.55	−0.413	−18.950	−2.552	0
0.6	0.578	−2.966	−3.179	1	0.6	0.933	−1.753	−3.117	1	0.6	0.869	−2.366	−2.791	1	0.6	−0.413	−18.254	−2.310	0
0.65	0.557	−3.012	−3.172	1	0.65	0.978	−0.621	−3.008	1	0.65	0.863	−2.892	−2.646	0	0.65	−0.305	−15.958	−2.310	0
0.7	0.648	−2.196	−3.225	1	0.7	1.000	−0.001	−2.913	1	0.7	0.845	−3.637	−2.901	0	0.7	−0.305	−14.943	−2.310	0
0.75	0.659	−2.054	−3.129	1	0.75	0.998	−0.048	−2.962	1	0.75	0.841	−3.654	−2.310	0	0.75	−0.305	−30.486	−2.444	0
0.8	0.689	−1.819	−3.112	1	0.8	1.025	0.456	−3.011	1	0.8	0.802	−5.056	−2.310	0	0.8	−0.305	86.904	−2.310	1
KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root
		8.89	3.00	0			6.7	3.0	0			5.63	3.02	0			86.90	3.03	0

Notes: The table shows point estimates, t-statistics (t-stat) and critical values (CV) for the 5 percent significance level. For the pointwise quantile unit root test, the null hypothesis ρ(τ) = 1 is rejected at the 5 percent level if the t-stat is numerically smaller than the CV. The last row for each regime reports results for the Kolmogorov–Smirnov (KS) test, which evaluates the joint null across all quantiles. The t-stat column displays the KS test statistic, and the CV column shows the corresponding critical value. The unit root null is rejected over the entire conditional distribution if the KS statistic exceeds the critical value.

Table D.6:

Quantilewise results for CPII-IW-food and CPII-IW-fuel.

Monetary targeting regime					Multiple indicators regime					IT regime					Pre-COVID IT regime
Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root
CPII-IW-food
0.1	0.808	−2.563	−2.446	0	0.1	0.855	−2.003	−2.558	1	0.1	0.544	−4.423	−2.586	0	0.1	−0.83	13.82	−2.44	1
0.15	0.832	−2.698	−3.113	1	0.15	0.856	−2.569	−2.684	1	0.15	0.672	−3.250	−3.037	0	0.15	−0.83	15.66	−2.31	1
0.2	0.820	−3.030	−3.024	0	0.2	0.855	−3.646	−2.698	0	0.2	0.850	−1.532	−2.865	1	0.2	−0.83	20.93	−2.31	1
0.25	0.834	−2.746	−2.954	1	0.25	0.874	−3.940	−2.829	0	0.25	0.814	−1.922	−2.593	1	0.25	−0.83	−13.03	−2.31	0
0.3	0.857	−2.609	−3.277	1	0.3	0.854	−4.351	−2.792	0	0.3	0.821	−1.652	−2.863	1	0.3	−0.83	−13.03	−2.31	0
0.35	0.854	−3.036	−3.099	1	0.35	0.842	−4.613	−3.104	0	0.35	0.810	−1.648	−2.919	1	0.35	−0.83	−11.22	−2.31	0
0.4	0.835	−3.290	−3.125	0	0.4	0.842	−4.143	−3.112	0	0.4	0.808	−1.765	−2.693	1	0.4	−0.96	−12.64	−2.31	0
0.45	0.805	−3.441	−3.207	0	0.45	0.863	−3.590	−3.106	0	0.45	0.804	−2.823	−2.726	0	0.45	−0.96	−13.12	−2.31	0
0.5	0.828	−2.851	−3.283	1	0.5	0.898	−2.363	−3.166	1	0.5	0.796	−2.574	−2.940	1	0.5	−1.03	−13.83	−2.31	0
0.55	0.832	−2.616	−3.220	1	0.55	0.909	−2.167	−3.310	1	0.55	0.817	−2.342	−2.729	1	0.55	−1.03	−13.64	−2.31	0
0.6	0.838	−2.504	−3.105	1	0.6	0.913	−1.833	−3.304	1	0.6	0.829	−1.918	−2.989	1	0.6	−0.96	−12.63	−2.31	0
0.65	0.849	−2.165	−3.214	1	0.65	0.936	−1.336	−3.291	1	0.65	0.883	−1.113	−2.959	1	0.65	−0.96	−11.96	−2.31	0
0.7	0.884	−1.572	−3.257	1	0.7	0.957	−0.864	−3.077	1	0.7	0.898	−0.894	−2.803	1	0.7	−0.96	−12.32	NaN	0
0.75	0.884	−1.427	−3.023	1	0.75	0.986	−0.239	−2.890	1	0.75	0.896	−0.947	−2.799	1	0.75	−0.96	−41.37	−2.44	0
0.8	0.852	−1.895	−2.988	1	0.8	0.994	−0.097	−2.760	1	0.8	0.932	−0.693	−3.398	1	0.8	−0.96	18.08	−2.31	1
KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV		KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root
		3.44	3.01	0			4.613	2.995	0			4.42	3.01	0			41.37	3.13	0
CPII-IW-fuel
0.1	0.48	−1.32	−3.09	1	0.1	0.890	−2.323	−2.705	1	0.1	0.820	−1.206	−2.310	1	0.1	−2.09	7.34	NaN	0
0.15	0.56	−2.31	−3.29	1	0.15	0.948	−1.026	−2.739	1	0.15	0.975	−0.178	−2.778	1	0.15	−2.64	10.42	−2.75	1
0.2	0.80	−1.29	−3.41	1	0.2	0.950	−1.256	−2.789	1	0.2	0.984	−0.118	−2.726	1	0.2	−2.09	−51.58	−3.13	0
0.25	0.79	−1.60	−3.37	1	0.25	0.942	−1.694	−2.951	1	0.25	0.967	−0.262	−2.310	1	0.25	−2.09	−7.07	−2.75	0
0.3	0.78	−1.86	−3.36	1	0.3	0.932	−2.504	−2.836	1	0.3	1.002	0.015	−2.944	1	0.3	−2.09	−7.15	−2.78	0
0.35	0.79	−1.85	−3.31	1	0.35	0.937	−2.437	−2.889	1	0.35	1.017	0.384	−2.829	1	0.35	−2.09	−6.82	−2.31	0
0.4	0.81	−1.97	−3.14	1	0.4	0.934	−2.716	−2.989	1	0.4	1.017	0.697	−2.310	1	0.4	−3.09	−9.53	−2.31	0
0.45	0.81	−3.03	−3.22	1	0.45	0.950	−2.290	−2.945	1	0.45	1.018	0.979	−2.392	1	0.45	−2.86	−9.32	−2.31	0
0.5	0.82	−3.15	−3.05	0	0.5	0.951	−1.996	−2.885	1	0.5	1.026	1.511	−2.310	1	0.5	−2.86	−9.45	−2.31	0
0.55	0.81	−3.65	−3.03	0	0.55	0.940	−2.400	−2.849	1	0.55	1.039	2.289	−2.310	1	0.55	−2.89	−9.39	−2.74	0
0.6	0.81	−3.06	−3.11	1	0.6	0.935	−2.620	−2.968	1	0.6	1.039	4.215	−2.310	1	0.6	−2.64	−8.47	−2.31	0
0.65	0.77	−2.89	−3.26	1	0.65	0.930	−2.398	−3.101	1	0.65	1.034	2.899	−2.310	1	0.65	−2.64	−8.02	−2.31	0
0.7	0.73	−2.68	−3.07	1	0.7	0.953	−1.359	−2.923	1	0.7	1.032	2.269	−2.310	1	0.7	−2.64	−46.27	−2.31	0
0.75	0.61	−2.46	−3.13	1	0.75	0.958	−1.180	−2.964	1	0.75	1.032	2.587	−2.310	1	0.75	−2.64	−42.91	−2.31	0
0.8	0.62	−1.96	−3.34	1	0.8	0.950	−1.330	−2.917	1	0.8	1.009	0.042	−2.310	1	0.8	−2.64	15.81	−3.00	1
KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root
		3.65	2.92	0			2.91	3.00	1			10.26	3.02	0			51.579	9.997	0

Notes: The table shows point estimates, t-statistics (t-stat) and critical values (CV) for the 5 percent significance level. For the pointwise quantile unit root test, the null hypothesis ρ(τ) = 1 is rejected at the 5 percent level if the t-stat is numerically smaller than the CV. The last row for each regime reports results for the Kolmogorov–Smirnov (KS) test, which evaluates the joint null across all quantiles. The t-stat column displays the KS test statistic, and the CV column shows the corresponding critical value. The unit root null is rejected over the entire conditional distribution if the KS statistic exceeds the critical value.

Table D.7:

Quantilewise results for CPII-IW-housing and CPII-IW-tobacco.

Monetary targeting regime					Multiple indicators regime					IT regime					Pre-COVID IT regime
Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root
CPII-IW-housing
0.1	0.714	−1.895	−2.590	1	0.1	0.804	−2.328	−2.560	1	0.1	0.825	−0.860	−2.915	1	0.1	−0.487	134.435	−3.167	1
0.15	0.910	−0.918	−2.328	1	0.15	0.996	−0.068	−2.374	1	0.15	0.836	−0.661	−2.729	1	0.15	−0.487	152.264	−2.857	1
0.2	0.950	−0.598	−2.310	1	0.2	0.998	−0.030	−2.615	1	0.2	0.802	−0.873	−2.764	1	0.2	−0.487	269.528	−2.465	1
0.25	0.972	−0.666	−2.310	1	0.25	0.999	−0.029	−2.585	1	0.25	1.001	0.005	−2.589	1	0.25	−0.487	−483.705	−2.310	0
0.3	0.975	−0.965	−2.310	1	0.3	1.000	−0.124	−2.651	1	0.3	0.999	−0.005	−2.310	1	0.3	−0.487	−150.501	−2.310	0
0.35	0.979	−1.189	−2.310	1	0.35	1.000	−0.047	−2.578	1	0.35	1.005	0.079	−2.310	1	0.35	−0.487	−160.716	−2.310	0
0.4	0.981	−2.147	−2.310	1	0.4	0.999	−0.501	−2.584	1	0.4	1.000	−0.008	−2.608	1	0.4	−0.502	−94.602	−2.600	0
0.45	0.986	−2.207	−2.310	1	0.45	0.999	−0.458	−2.554	1	0.45	0.980	−0.362	−2.310	1	0.45	−0.502	−98.209	−2.310	0
0.5	0.988	−1.714	−2.515	1	0.5	1.000	−0.033	−2.310	1	0.5	0.979	−2.106	−2.310	1	0.5	−0.502	−99.548	−2.310	0
0.55	0.989	−1.561	−2.504	1	0.55	1.000	−0.121	−2.351	1	0.55	0.979	−2.214	−2.310	1	0.55	−0.534	−100.315	−2.310	0
0.6	0.992	−1.176	−2.371	1	0.6	1.000	−0.122	−2.410	1	0.6	0.982	−1.225	−2.310	1	0.6	−0.534	−96.631	−2.310	0
0.65	0.994	−0.742	−2.454	1	0.65	1.000	−0.189	−2.517	1	0.65	0.984	−0.991	−2.310	1	0.65	−0.534	−91.505	−2.469	0
0.7	0.995	−0.233	−2.345	1	0.7	0.999	−0.733	−2.451	1	0.7	0.984	−1.107	−2.310	1	0.7	−0.534	−101.929	−2.310	0
0.75	0.988	−0.391	−2.440	1	0.75	0.998	−0.185	−2.363	1	0.75	0.984	−0.392	−2.310	1	0.75	−0.534	−94.608	−2.310	0
0.8	0.993	−0.160	−2.491	1	0.8	0.997	−0.114	−2.331	1	0.8	0.986	−0.313	−2.310	1	0.8	−0.534	513.510	−3.305	1
KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV		KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root
		7.925	3.008	0			5.303	2.992	0			8.755	3.027	0			566.101	6.677
CPII-IW-tobacco
0.1	0.752	−4.677	−2.380	0	0.1	0.921	−1.911	−2.618	1	0.1	0.736	−10.443	−2.310	0	0.1	0.378	2.523	−2.792	1
0.15	0.823	−3.291	−2.659	0	0.15	0.951	−1.344	−2.859	1	0.15	0.718	−4.512	−2.310	0	0.15	0.378	6.560	−3.155	1
0.2	0.855	−3.246	−2.501	0	0.2	0.952	−1.610	−2.866	1	0.2	0.732	−4.355	−2.410	0	0.2	0.378	7.232	−3.243	1
0.25	0.872	−3.355	−2.781	0	0.25	0.950	−2.220	−2.923	1	0.25	0.795	−2.527	−2.847	1	0.25	0.378	−6.067	−2.310	0
0.3	0.890	−3.153	−2.953	0	0.3	0.949	−2.212	−3.023	1	0.3	0.828	−2.066	−2.800	1	0.3	0.000	−3.492	−2.310	0
0.35	0.907	−2.545	−3.160	1	0.35	0.961	−1.748	−3.132	1	0.35	0.786	−2.439	−2.784	1	0.35	0.000	−3.576	−2.310	0
0.4	0.898	−2.955	−3.219	1	0.4	0.972	−1.270	−3.263	1	0.4	0.826	−2.076	−2.554	1	0.4	0.000	−3.776	−2.310	0
0.45	0.907	−2.528	−3.226	1	0.45	0.974	−1.261	−3.160	1	0.45	0.828	−2.105	−2.898	1	0.45	0.000	−3.920	−2.310	0
0.5	0.921	−2.152	−3.149	1	0.5	0.968	−1.614	−3.086	1	0.5	0.854	−2.076	−2.667	1	0.5	−0.029	−4.090	−2.310	0
0.55	0.938	−1.458	−3.069	1	0.55	0.965	−1.576	−3.015	1	0.55	0.852	−2.509	−2.692	1	0.55	−0.133	−4.441	−2.310	0
0.6	0.947	−1.255	−3.211	1	0.6	0.966	−1.521	−2.914	1	0.6	0.854	−2.334	−2.527	1	0.6	−0.105	−4.171	−2.328	0
0.65	0.966	−0.789	−3.280	1	0.65	0.966	−1.311	−2.882	1	0.65	0.840	−2.503	−2.587	1	0.65	−0.105	−5.797	−2.310	0
0.7	0.942	−1.243	−3.098	1	0.7	0.952	−1.641	−3.008	1	0.7	0.846	−2.208	−2.407	1	0.7	−0.105	−5.428	−2.404	0
0.75	0.954	−0.703	−3.010	1	0.75	0.939	−1.903	−2.837	1	0.75	0.849	−1.503	−2.310	1	0.75	−0.105	−12.310	−2.732	0
0.8	0.946	−0.836	−3.045	1	0.8	0.927	−1.960	−2.921	1	0.8	0.877	−1.208	−2.802	1	0.8	−0.105	5.809	−3.398	1
KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root
		4.677	3.002	0			2.342	3.009	1			10.443	3.013	0			12.310	3.306	0

Notes: The table shows point estimates, t-statistics (t-stat) and critical values (CV) for the 5 percent significance level. For the pointwise quantile unit root test, the null hypothesis ρ(τ) = 1 is rejected at the 5 percent level if the t-stat is numerically smaller than the CV. The last row for each regime reports results for the Kolmogorov–Smirnov (KS) test, which evaluates the joint null across all quantiles. The t-stat column displays the KS test statistic, and the CV column shows the corresponding critical value. The unit root null is rejected over the entire conditional distribution if the KS statistic exceeds the critical value.

Table D.8:

Quantilewise results for CPII-IW-miscellaneous and CPII-IW-cloth.

Monetary targeting regime					Multiple indicators regime					IT regime					Pre-COVID IT regime
Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root
CPII-IW-miscellaneous
0.1	0.527	−2.399	−2.936	1	0.1	0.923	−1.684	−2.611	1	0.1	0.769	−4.371	−3.187	0	0.1	−0.176	10.066	−2.310	1
0.15	0.721	−1.607	−3.321	1	0.15	0.862	−2.869	−2.789	0	0.15	0.754	−4.047	−3.234	0	0.15	−0.175	11.400	−2.310	1
0.2	0.751	−2.562	−2.885	1	0.2	0.891	−2.491	−3.025	1	0.2	0.792	−2.911	−3.314	1	0.2	−0.176	19.385	−2.310	1
0.25	0.755	−3.190	−2.946	0	0.25	0.875	−3.012	−3.045	1	0.25	0.790	−2.481	−2.875	1	0.25	−0.176	−23.766	−2.310	0
0.3	0.764	−4.184	−2.978	0	0.3	0.898	−2.580	−3.007	1	0.3	0.818	−1.592	−3.153	1	0.3	−0.176	−7.541	−2.679	0
0.35	0.801	−3.759	−3.107	0	0.35	0.939	−1.743	−3.041	1	0.35	0.787	−1.812	−3.161	1	0.35	−0.179	−8.076	−2.310	0
0.4	0.805	−4.125	−2.958	0	0.4	0.958	−1.397	−2.998	1	0.4	0.841	−1.421	−3.127	1	0.4	−0.200	−8.683	−2.310	0
0.45	0.805	−4.858	−3.156	0	0.45	0.960	−1.303	−3.011	1	0.45	0.844	−1.666	−2.884	1	0.45	−0.200	−9.014	−2.310	0
0.5	0.807	−3.980	−3.190	0	0.5	0.966	−0.946	−3.085	1	0.5	0.819	−1.915	−2.895	1	0.5	−0.206	−9.178	−2.310	0
0.55	0.762	−4.522	−3.167	0	0.55	0.965	−1.122	−3.093	1	0.55	0.812	−1.842	−3.209	1	0.55	−0.053	−7.907	−2.310	0
0.6	0.768	−4.223	−3.071	0	0.6	0.960	−1.279	−3.012	1	0.6	0.813	−1.850	−2.963	1	0.6	−0.118	−8.084	−2.310	0
0.65	0.762	−3.835	−3.002	0	0.65	0.960	−1.207	−3.137	1	0.65	0.784	−2.114	−2.682	1	0.65	−0.118	−7.682	−2.310	0
0.7	0.769	−2.699	−3.274	1	0.7	0.953	−1.307	−3.124	1	0.7	0.768	−1.981	−2.631	1	0.7	−0.118	−8.744	−2.310	0
0.75	0.821	−2.070	−2.923	1	0.75	0.984	−0.413	−3.136	1	0.75	0.787	−2.375	−2.929	1	0.75	−0.118	−9.429	−2.310	0
0.8	0.842	−1.209	−2.989	1	0.8	0.977	−0.506	−3.083	1	0.8	0.805	−2.173	−2.893	1	0.8	−0.118	12.556	−2.310	1
KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV		KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root
		5.24	3.00	0			4.72	3.00	0			4.37	3.01	0			23.766	3.022
CPII-IW-cloth
0.1	0.717	−2.903	−2.751	0	0.1	0.890	−1.065	−2.768	1	0.1	0.944	−0.143	−3.012	1	0.1	−4.954	6.710	−2.310	1
0.15	0.787	−2.493	−2.814	1	0.15	0.924	−1.272	−2.573	1	0.15	0.972	−0.079	−3.012	1	0.15	−4.954	7.600	−2.310	1
0.2	0.788	−2.607	−3.030	1	0.2	0.928	−1.990	−2.923	1	0.2	0.970	−0.095	−2.745	1	0.2	−4.954	10.294	−2.310	1
0.25	0.860	−2.124	−2.760	1	0.25	0.930	−2.596	−2.822	1	0.25	0.976	−0.748	−2.435	1	0.25	−4.954	−48.912	−2.315	0
0.3	0.870	−2.526	−2.794	1	0.3	0.949	−2.198	−2.748	1	0.3	0.974	−1.196	−2.713	1	0.3	−4.954	−7.455	−2.310	0
0.35	0.885	−2.264	−2.895	1	0.35	0.957	−2.042	−2.614	1	0.35	0.976	−1.069	−2.476	1	0.35	−4.954	−5.729	−2.310	0
0.4	0.922	−1.878	−2.698	1	0.4	0.965	−1.614	−2.824	1	0.4	0.980	−0.840	−2.310	1	0.4	−4.954	−4.546	−2.310	0
0.45	0.926	−1.754	−2.846	1	0.45	0.971	−1.382	−2.727	1	0.45	0.989	−0.549	−2.310	1	0.45	−4.648	−4.477	−2.310	0
0.5	0.935	−1.621	−2.964	1	0.5	0.977	−1.145	−2.898	1	0.5	0.994	−0.305	−2.436	1	0.5	−4.648	−4.538	−2.310	0
0.55	0.963	−0.940	−2.926	1	0.55	0.973	−1.221	−2.933	1	0.55	1.023	1.207	−2.395	1	0.55	−4.843	−4.632	−2.310	0
0.6	0.970	−0.774	−3.117	1	0.6	0.971	−1.180	−3.003	1	0.6	1.020	0.996	−2.370	1	0.6	−1.335	−1.783	−2.314	1
0.65	0.964	−0.792	−2.955	1	0.65	0.975	−0.921	−2.984	1	0.65	1.002	0.096	−2.389	1	0.65	−1.335	−1.689	−2.310	1
0.7	0.959	−0.774	−2.830	1	0.7	0.976	−0.918	−2.922	1	0.7	1.007	0.054	−2.383	1	0.7	−1.335	−1.712	−2.310	1
0.75	0.943	−0.682	−2.967	1	0.75	0.957	−1.355	−2.930	1	0.75	1.005	0.040	−2.418	1	0.75	0.090	−0.619	−2.340	1
0.8	0.962	−0.437	−2.983	1	0.8	0.996	−0.085	−2.916	1	0.8	1.009	0.067	−2.537	1	0.8	0.090	1.545	−2.310	1
KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root
		2.95	3.02	1			2.60	3.00	1			33.20	4.91	0			48.91	3.00	0

Notes: The table shows point estimates, t-statistics (t-stat) and critical values (CV) for the 5 percent significance level. For the pointwise quantile unit root test, the null hypothesisρ(τ) = 1 is rejected at the 5 percent level if the t-stat is numerically smaller than the CV. The last row for each regime reports results for the Kolmogorov–Smirnov (KS) test, which evaluates the joint null across all quantiles. The t-stat column displays the KS test statistic, and the CV column shows the corresponding critical value. The unit root null is rejected over the entire conditional distribution if the KS statistic exceeds the critical value.

Table D.9:

Quantilewise results for WPII and WPII-core.

Multiple indicators regime					IT regime					Pre-COVID IT regime
Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root
WPII
0.1	0.944	−1.644	−2.591	1	0.1	0.829	−8.913	−2.505	0	0.1	0.521	1.827	−2.310	1
0.15	0.962	−1.244	−2.818	1	0.15	0.862	−4.125	−2.868	0	0.15	0.521	2.070	−2.310	1
0.2	0.960	−1.457	−2.762	1	0.2	0.864	−3.311	−2.907	0	0.2	0.521	2.282	−2.310	1
0.25	0.977	−0.770	−3.034	1	0.25	0.862	−2.618	−2.500	0	0.25	0.521	−2.891	−2.310	0
0.3	0.969	−1.037	−2.998	1	0.3	0.857	−2.428	−2.894	1	0.3	0.521	−2.004	−2.310	1
0.35	0.949	−1.880	−3.033	1	0.35	0.834	−2.631	−2.854	1	0.35	0.521	−1.320	−2.340	1
0.4	0.949	−1.986	−3.008	1	0.4	0.847	−2.427	−3.172	1	0.4	0.521	−1.394	−2.681	1
0.45	0.955	−1.837	−2.974	1	0.45	0.869	−2.146	−2.953	1	0.45	0.320	−2.052	−2.399	1
0.5	0.955	−1.793	−2.797	1	0.5	0.821	−2.681	−2.993	1	0.5	−0.503	−4.600	−2.310	0
0.55	0.953	−1.992	−2.921	1	0.55	0.818	−2.731	−2.976	1	0.55	−0.656	−4.999	−2.310	0
0.6	0.953	−1.986	−3.010	1	0.6	0.813	−2.792	−3.195	1	0.6	−0.832	−5.328	−2.450	0
0.65	0.943	−2.503	−3.064	1	0.65	0.776	−3.019	−2.873	0	0.65	−0.660	−4.572	−2.544	0
0.7	0.948	−2.586	−3.032	1	0.7	0.774	−3.457	−2.728	0	0.7	−0.660	−4.281	−2.310	0
0.75	0.949	−2.352	−3.064	1	0.75	0.784	−3.487	−2.664	0	0.75	−0.660	−7.087	−2.310	0
0.8	0.954	−2.025	−2.910	1	0.8	0.810	−3.625	−2.963	0	0.8	−0.660	10.663	−2.672	1
KS-test		t-stat	CV		KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root
		2.59	3.01	1			8.91	4.27	0			10.66	5.79	0
WPII-core
0.1	0.909	−2.561	−2.921	1	0.1	0.765	−9.922	−2.775	0	0.1	0.564	4.608	−2.310	1
0.15	0.929	−2.101	−3.091	1	0.15	0.779	−9.923	−3.318	0	0.15	0.564	5.219	−3.240	1
0.2	0.916	−2.487	−3.127	1	0.2	0.790	−6.977	−2.646	0	0.2	0.564	5.754	−2.885	1
0.25	0.926	−2.329	−3.002	1	0.25	0.793	−5.831	−2.418	0	0.25	0.564	−19.291	−2.616	0
0.3	0.942	−2.028	−3.125	1	0.3	0.802	−4.155	−2.902	0	0.3	0.564	−20.776	−2.431	0
0.35	0.950	−1.901	−3.106	1	0.35	0.807	−3.961	−2.763	0	0.35	0.564	−5.757	−2.797	0
0.4	0.963	−1.527	−3.109	1	0.4	0.801	−3.665	−2.977	0	0.4	0.564	−6.080	−2.501	0
0.45	0.955	−1.841	−3.177	1	0.45	0.806	−3.726	−2.833	0	0.45	0.550	−2.628	−2.435	0
0.5	0.957	−1.910	−3.157	1	0.5	0.823	−3.086	−3.023	0	0.5	0.620	−2.247	−2.310	1
0.55	0.962	−1.623	−3.225	1	0.55	0.851	−2.579	−3.064	1	0.55	0.620	−2.217	−2.310	1
0.6	0.951	−2.015	−3.131	1	0.6	0.854	−2.488	−2.914	1	0.6	0.640	−2.023	−2.310	1
0.65	0.930	−3.238	−3.025	0	0.65	0.838	−2.923	−2.960	1	0.65	0.692	−1.639	−2.310	1
0.7	0.952	−1.708	−2.954	1	0.7	0.834	−2.732	−2.849	1	0.7	0.692	−1.535	−2.310	1
0.75	0.942	−2.011	−2.949	1	0.75	0.816	−4.188	−2.897	0	0.75	0.805	−1.010	−2.310	1
0.8	0.955	−1.281	−3.154	1	0.8	0.827	−4.995	−3.030	0	0.8	0.805	−3.819	−2.385	0
KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root
		6.54	3.00	0			9.92	3.01	0			20.78	9.92	0

Notes: The table shows point estimates, t-statistics (t-stat) and critical values (CV) for the 5 percent significance level. For the pointwise quantile unit root test, the null hypothesis ρ(τ) = 1 is rejected at the 5 percent level if the t-stat is numerically smaller than the CV. The last row for each regime reports results for the Kolmogorov–Smirnov (KS) test, which evaluates the joint null across all quantiles. The t-stat column displays the KS test statistic, and the CV column shows the corresponding critical value. The unit root null is rejected over the entire conditional distribution if the KS statistic exceeds the critical value.

Table D.10:

Quantilewise results for WPII-primary and WPII-fuel.

Multiple indicators regime					IT regime					Pre-COVID IT regime
Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root
WPII-primary
0.1	0.813	−2.797	−2.734	0	0.1	0.399	−43.438	−2.828	0	0.1	−1.311	8.924	−2.310	1
0.15	0.817	−3.300	−2.699	0	0.15	0.377	−12.496	−2.855	0	0.15	−1.311	10.108	−3.252	1
0.2	0.874	−3.189	−2.837	0	0.2	0.382	−5.842	−3.082	0	0.2	−1.311	11.143	−2.902	1
0.25	0.877	−3.667	−2.751	0	0.25	0.411	−4.078	−3.087	0	0.25	−1.311	−8.907	−2.601	0
0.3	0.892	−3.288	−2.795	0	0.3	0.407	−4.174	−3.215	0	0.3	−1.311	−6.561	−2.501	0
0.35	0.896	−2.846	−2.932	1	0.35	0.499	−3.632	−3.321	0	0.35	−1.311	−7.007	−2.310	0
0.4	0.899	−3.036	−3.139	1	0.4	0.538	−3.171	−3.301	1	0.4	−1.311	−7.400	−2.310	0
0.45	0.877	−3.561	−3.173	0	0.45	0.534	−2.988	−3.316	1	0.45	−1.326	−7.732	−2.310	0
0.5	0.878	−3.819	−3.258	0	0.5	0.534	−2.752	−3.369	1	0.5	−0.376	−4.638	−2.310	0
0.55	0.885	−3.216	−3.187	0	0.55	0.533	−2.947	−2.973	1	0.55	−0.376	−4.576	−2.310	0
0.6	0.894	−2.415	−3.159	1	0.6	0.526	−2.947	−2.782	0	0.6	−0.803	−5.772	−2.310	0
0.65	0.915	−1.924	−3.103	1	0.65	0.460	−2.980	−2.795	0	0.65	−0.803	−5.466	−2.530	0
0.7	0.934	−1.553	−3.106	1	0.7	0.555	−2.142	−2.683	1	0.7	−0.803	−5.118	−2.310	0
0.75	0.944	−1.303	−3.003	1	0.75	0.532	−2.184	−2.887	1	0.75	−0.803	−15.035	−2.310	0
0.8	1.001	0.016	−3.098	1	0.8	0.515	−2.164	−2.761	1	0.8	−0.803	8.794	−2.310	1
KS-test		t-stat	CV		KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root
		3.82	3.00	0			43.44	2.98	0			15.03	3.02	0
WPII-fuel
0.1	0.902	−2.082	−3.104	1	0.1	0.831	−3.334	−2.839	0	0.1	−0.374	11.419	−2.310	1
0.15	0.869	−2.482	−2.999	1	0.15	0.943	−1.080	−3.292	1	0.15	−0.374	12.934	−2.310	1
0.2	0.883	−2.427	−3.101	1	0.2	0.944	−0.802	−2.547	1	0.2	−0.374	14.303	−2.344	1
0.25	0.907	−2.016	−3.003	1	0.25	0.946	−0.761	−2.555	1	0.25	−0.374	−19.034	−2.310	0
0.3	0.919	−1.885	−3.070	1	0.3	0.943	−0.781	−2.955	1	0.3	−0.374	−20.501	−2.310	0
0.35	0.936	−1.660	−2.905	1	0.35	0.901	−1.342	−3.103	1	0.35	−0.374	−7.497	−2.310	0
0.4	0.931	−2.137	−3.102	1	0.4	0.864	−2.144	−2.923	1	0.4	−0.374	−7.919	−2.310	0
0.45	0.923	−2.700	−2.988	1	0.45	0.863	−2.431	−2.999	1	0.45	−0.202	−7.192	−2.310	0
0.5	0.923	−2.828	−3.013	1	0.5	0.846	−2.661	−2.956	1	0.5	−0.202	−7.290	−2.310	0
0.55	0.923	−2.749	−2.981	1	0.55	0.852	−2.108	−2.754	1	0.55	−0.202	−7.192	−2.390	0
0.6	0.937	−2.169	−2.906	1	0.6	0.845	−2.335	−2.786	1	0.6	−0.202	−6.928	−2.473	0
0.65	0.925	−2.648	−2.899	1	0.65	0.842	−2.330	−2.947	1	0.65	−0.203	−6.563	−2.607	0
0.7	0.918	−2.674	−2.979	1	0.7	0.846	−2.190	−2.546	1	0.7	−0.098	−5.617	−2.929	0
0.75	0.929	−1.973	−2.744	1	0.75	0.817	−2.640	−2.487	0	0.75	−0.098	−7.920	−2.328	0
0.8	0.931	−1.728	−2.861	1	0.8	0.829	−2.587	−2.556	0	0.8	−0.098	−61.345	−2.538	0
KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root
		5.495	3.016	0			3.847	3.029	0			61.35	3.01	0

Notes: The table shows point estimates, t-statistics (t-stat) and critical values (CV) for the 5 percent significance level. For the pointwise quantile unit root test, the null hypothesis ρ(τ) = 1 is rejected at the 5 percent level if the t-stat is numerically smaller than the CV. The last row for each regime reports results for the Kolmogorov–Smirnov (KS) test, which evaluates the joint null across all quantiles. The t-stat column displays the KS test statistic, and the CV column shows the corresponding critical value. The unit root null is rejected over the entire conditional distribution if the KS statistic exceeds the critical value.

Table D.11:

Quantilewise results for WPII-manufacturing.

Multiple indicators regime					IT regime					Pre-COVID IT regime
Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root	Quantiles	ρ(τ)	t-stat	CV	Unit root
WPII-manufacturing
0.1	0.920	−2.203	−2.800	1	0.1	0.907	−2.442	−2.867	1	0.1	0.006	4.234	−2.597	1
0.15	0.928	−1.863	−2.835	1	0.15	0.860	−3.421	−2.972	0	0.15	0.006	4.795	−2.310	1
0.2	0.928	−2.102	−3.109	1	0.2	0.870	−3.108	−2.862	0	0.2	0.006	13.579	−2.627	1
0.25	0.942	−2.162	−3.025	1	0.25	0.871	−2.572	−2.828	1	0.25	0.006	−4.729	−2.310	0
0.3	0.945	−1.850	−2.969	1	0.3	0.886	−2.211	−2.966	1	0.3	0.006	−4.366	−2.310	0
0.35	0.952	−1.652	−3.045	1	0.35	0.898	−1.560	−2.932	1	0.35	0.117	−4.141	−2.310	0
0.4	0.950	−1.873	−3.163	1	0.4	0.878	−1.923	−3.179	1	0.4	0.117	−3.061	−2.310	0
0.45	0.949	−2.292	−3.239	1	0.45	0.869	−2.488	−2.680	1	0.45	0.117	−3.178	−2.310	0
0.5	0.957	−2.027	−3.113	1	0.5	0.877	−2.475	−3.282	1	0.5	−0.143	−4.171	−2.310	0
0.55	0.951	−2.304	−3.162	1	0.55	0.903	−1.872	−3.061	1	0.55	−0.143	−4.115	−2.482	0
0.6	0.953	−2.201	−3.009	1	0.6	0.867	−2.490	−2.719	1	0.6	−0.154	−4.002	−3.097	0
0.65	0.960	−1.991	−2.928	1	0.65	0.863	−2.841	−2.927	1	0.65	−0.154	−3.790	−2.806	0
0.7	0.967	−1.758	−2.946	1	0.7	0.861	−2.522	−2.604	1	0.7	−0.692	−7.449	−2.429	0
0.75	0.971	−1.184	−2.873	1	0.75	0.870	−2.646	−2.761	1	0.75	−0.647	−11.754	−2.310	0
0.8	0.977	−0.831	−2.815	1	0.8	0.856	−4.117	−2.686	0	0.8	−0.647	20.972	NaN	1
KS-test		t-stat	CV		KS-test		t-stat	CV	Unit root	KS-test		t-stat	CV	Unit root
		2.695	3.012	1			4.117	3.020	0			20.97	5.16	0

Notes: The table shows point estimates, t-statistics (t-stat) and critical values (CV) for the 5 percent significance level. For the pointwise quantile unit root test, the null hypothesis ρ(τ) = 1 is rejected at the 5 percent level if the t-stat is numerically smaller than the CV. The last row for each regime reports results for the Kolmogorov–Smirnov (KS) test, which evaluates the joint null across all quantiles. The t-stat column displays the KS test statistic, and the CV column shows the corresponding critical value. The unit root null is rejected over the entire conditional distribution if the KS statistic exceeds the critical value.

Figure D.1:

CPI-IW and its disaggregates inflation persistence. Notes: The graph shows inflation persistence of CPI-IW, and its disaggregates considering seven years rolling estimates of ρ at the conditional mean (solid line) together with 95 percent bootstrapped confidence band (orange areas).

Figure D.2:

WPI and its disaggregates inflation persistence. Notes: The graph shows inflation persistence of WPI, and its disaggregates considering seven years rolling estimates of ρ at the conditional mean (solid line) together with 95 percent bootstrapped confidence band (orange areas).

Figure D.3:

QQ plots for CPII-IW and its disaggregates. Red points denote the quantiles of the data series and blue lines represent the theoretical quantiles of a normal distribution.

Figure D.4:

QQ plots for WPI and its disaggregates. Red points denote the quantiles of the data series and blue lines represent the theoretical quantiles of a normal distribution.

Figure D.5:

QQ plots for CPI-IW and its disaggregates persistence. Red points denote the quantiles of the data series and blue lines represent the theoretical quantiles of a normal distribution.

Table D.12:

ADF test for unit root in M3 growth across monetary regimes.

Regime	Z(t)	1 % CV	5 % CV	10 % CV
Monetary targeting	−13.022	−3.495	−2.887	−2.577
Multiple indicators	−14.539	−3.470	−2.882	−2.572
Inflation targeting	−10.358	−3.525	−2.899	−2.584

References

Abakah, Emmanuel Joel Aikins, John W. Goodell, Zunaidah Sulong, and Mohammad Abdullah. 2024. “Wavelet Quantile Correlation Between DeFi Assets and Banking Stocks.” Finance Research Letters 70: 106272. https://doi.org/10.1016/j.frl.2024.106272.Search in Google Scholar

Anguyo, Francis Leni, Rangan Gupta, and Kevin Kotzé. 2020. “Inflation Dynamics in Uganda: A Quantile Regression Approach.” Macroeconomics and Finance in Emerging Market Economies 13 (2): 161–87. https://doi.org/10.1080/17520843.2019.1596963.Search in Google Scholar

Arsić, Milojko, Zorica Mladenović, and Aleksandra Nojković. 2022. “Macroeconomic Performance of Inflation Targeting in European and Asian Emerging Economies.” Journal of Policy Modeling 44 (3): 675–700. https://doi.org/10.1016/j.jpolmod.2022.06.002.Search in Google Scholar

Asl, Mahdi Ghaemi, Giorgio Canarella, Stephen M. Miller, and Hamid Reza Tavakkoli. 2023. “Does Real Interest Rate Parity Really Work? Historical Evidence from a Discrete Wavelet Perspective.” Studies in Nonlinear Dynamics & Econometrics 27 (4): 485–518. https://doi.org/10.1515/snde-2021-0067.Search in Google Scholar

Azad, Rohit, and Anupam Das. 2014. “Inflation Targeting in Developing Countries: Barking up the Wrong Tree.” In Market, Regulations and Finance: Global Meltdown and the Indian Economy, edited by Ratan Khasnabis, and Indrani Chakraborty, 95–111. New Delhi: Springer India.10.1007/978-81-322-1795-4_7Search in Google Scholar

Ball, Laurence, Anusha Chari, and Prachi Mishra. 2016. Understanding Inflation in India. Working Paper, Working Paper Series 22948. National Bureau of Economic Research.10.3386/w22948Search in Google Scholar

Barsky, Robert B. 1987. “The Fisher Hypothesis and the Forecastability and Persistence of Inflation.” Journal of Monetary Economics 19 (1): 3–24. https://doi.org/10.1016/0304-3932(87)90026-2.Search in Google Scholar

Batini, Nicoletta, and Edward Nelson. 2001. “The Lag from Monetary Policy Actions to Inflation: Friedman Revisited.” International Finance 4 (3): 381–400. https://doi.org/10.1111/1468-2362.00079.Search in Google Scholar

Beechey, Meredith, and Pär Österholm. 2009. “Time-Varying Inflation Persistence in the Euro Area.” Economic Modelling 26 (2): 532–5. https://doi.org/10.1016/j.econmod.2008.11.001.Search in Google Scholar

Behera, Harendra Kumar, and Michael Debabrata Patra. 2022. “Measuring Trend Inflation in India.” Journal of Asian Economics 80: 101474. https://doi.org/10.1016/j.asieco.2022.101474.Search in Google Scholar

Bhatt, Vipul, and N. Kundan Kishor. 2016. “Are All Movements in Food and Energy Prices Transitory? Evidence from India.” Journal of Policy Modeling 37 (1): 92–106.10.1016/j.jpolmod.2015.01.005Search in Google Scholar

Bouri, Elie, Remzi Gök, Eray Gemici, and Erkan Kara. 2024. “Do Geopolitical Risk, Economic Policy Uncertainty, and Oil Implied Volatility Drive Assets Across Quantiles and Time-Horizons?” The Quarterly Review of Economics and Finance 93: 137–54. https://doi.org/10.1016/j.qref.2023.12.004.Search in Google Scholar

Bratsiotis, George J., Jakob Madsen, and Christopher Martin. 2015. “Inflation Targeting and Inflation Persistence.” Economic and Political Studies 3 (1): 3–17. https://doi.org/10.1080/20954816.2015.11673835.Search in Google Scholar

Calvo, Guillermo A., and Frederic S. Mishkin. 2007. “The Mirage of Exchange-Rate Regimes for Emerging Market Countries.” In Monetary Policy Strategy. The MIT Press.10.7551/mitpress/7412.003.0025Search in Google Scholar

Çiçek and Akar, 2013 Çiçek, Serkan, and Cüneyt Akar. 2013. “The Asymmetry of Inflation Adjustment in Turkey.” Economic Modelling 31: 104–18. https://doi.org/10.1016/j.econmod.2012.11.026.Search in Google Scholar

Daniel Chiquiar, Antonio E. Noriega, and Manuel Ramos-Francia. 2010. “A Time-Series Approach to Test a Change in Inflation Persistence: The Mexican Experience.” Applied Economics 42 (24): 3067–75. https://doi.org/10.1080/00036840801982684.Search in Google Scholar

Darbha, Gangadhar, and Urjit R. Patel. 2012. Dynamics of Inflation Herding: Decoding India’s Inflationary Process. Brookings.Search in Google Scholar

Daubechies, Ingrid. 1992. Ten Lectures on Wavelets. SIAM.10.1137/1.9781611970104Search in Google Scholar

de Haan, Jakob, Sylvester C. W. Eijffinger, and Krzysztof Rybiński. 2007. “Central Bank Transparency and Central Bank Communication: Editorial Introduction.” European Journal of Political Economy 23 (1): 1–8. https://doi.org/10.1016/j.ejpoleco.2006.09.010.Search in Google Scholar

di Giovanni, Julian, and Jay C. Shambaugh. 2008. “The Impact of Foreign Interest Rates on the Economy: The Role of the Exchange Rate Regime.” Journal of International Economics 74 (2): 341–61. https://doi.org/10.1016/j.jinteco.2007.09.002.Search in Google Scholar

Dossche, Maarten, and Gerdie Everaert. 2005. Measuring Inflation Persistence: A Structural Time Series Approach. Technical report 70. NBB Working Paper.10.2139/ssrn.1690515Search in Google Scholar

Dua, Pami. 2020. “Monetary Policy Framework in India.” Indian Economic Review 55: 117–54. https://doi.org/10.1007/s41775-020-00085-3.Search in Google Scholar PubMed PubMed Central

Dua, Pami, and Deepika Goel. 2021a. “Determinants of Inflation in India.” The Journal of Developing Areas 55 (2). https://doi.org/10.1353/jda.2021.0040.Search in Google Scholar

Dua, Pami, and Deepika Goel. 2021b. “Inflation Persistence in India.” Journal of Quantitative Economics 19 (3): 525–53. https://doi.org/10.1007/s40953-021-00237-z.Search in Google Scholar

Erceg, Christopher J., and Andrew T. Levin. 2003. “Imperfect Credibility and Inflation Persistence.” Journal of Monetary Economics 50 (4): 915–44. https://doi.org/10.1016/S0304-3932(03)00036-9.Search in Google Scholar

Fraga, Arminio, Ilan Goldfajn, and André Minella. 2003. “Inflation Targeting in Emerging Market Economies.” NBER Macroeconomics Annual 18: 365–400. https://doi.org/10.1086/ma.18.3585264.Search in Google Scholar

Friedman, Milton. 2017. “Quantity Theory of Money.” In The New Palgrave Dictionary of Economics, 1–31. London: Palgrave Macmillan UK.10.1057/978-1-349-95121-5_1640-1Search in Google Scholar

Fuhrer, Jeffrey C. 2010. “Inflation Persistence.” In Handbook of Monetary Economics, Vol. 3, 423–86. Elsevier.10.1016/B978-0-444-53238-1.00009-0Search in Google Scholar

Gaglianone, Wagner Piazza, Osmani Teixeira de Carvalho Guillén, and Francisco Marcos Rodrigues Figueiredo. 2018. “Estimating Inflation Persistence by Quantile Autoregression with Quantile-Specific Unit Roots.” Economic Modelling 73: 407–30. https://doi.org/10.1016/j.econmod.2018.04.018.Search in Google Scholar

Gallegati, Marco, and Mauro Gallegati. 2007. “Wavelet Variance Analysis of Output in G-7 Countries.” Studies in Nonlinear Dynamics & Econometrics 11 (3). https://doi.org/10.2202/1558-3708.1435.Search in Google Scholar

Galvao, Antonio F. 2009. “Unit Root Quantile Autoregression Testing Using Covariates.” Journal of Econometrics 152 (2): 165–78. https://doi.org/10.1016/j.jeconom.2009.01.007.Search in Google Scholar

Gençay, Ramazan, Giuseppe Ballocchi, Michel Dacorogna, Richard Olsen, and Olivier Pictet. 2002. “Real-Time Trading Models and the Statistical Properties of Foreign Exchange Rates.” International Economic Review: 463–91. https://doi.org/10.1111/1468-2354.t01-2-00023.Search in Google Scholar

Gençay, Ramazan, Faruk Selçuk, and Brandon J. Whitcher. 2002. “7 – Wavelets for Variance-Covariance Estimation.” In An Introduction to Wavelets and Other Filtering Methods in Finance and Economics, edited by Ramazan Gençay, Faruk Selçuk, and Brandon J. Whitcher, 235–71. San Diego: Academic Press.10.1016/B978-012279670-8.50010-0Search in Google Scholar

Gerlach, Stefan, and Peter Tillmann. 2012. “Inflation Targeting and Inflation Persistence in Asia-Pacific.” Journal of Asian Economics 23 (4): 360–73. https://doi.org/10.1016/j.asieco.2012.03.002.Search in Google Scholar

Goyal, Ashima. 2015. “Understanding High Inflation Trend in India.” South Asian Journal of Macroeconomics and Public Finance 4 (1): 1–42. https://doi.org/10.1177/2277978715574614.Search in Google Scholar

Goyal, Ashima. 2022. “Flexible Inflation Targeting: Concepts and Application in India.” Indian Public Policy Review 3 (5): 1–21. https://doi.org/10.55763/ippr.2022.03.05.001.Search in Google Scholar

Goyal, Ashima, and Abhishek Kumar. 2024. “What Drives Indian Inflation? Demand or Supply.” In Practical Economic Analysis and Computation: A Festschrift in Honor of Professor Kirit Parikh, edited by Probal Pratap Ghosh, Rajbans Talwar, and Sureshbabu Syamasundar Velagapudi, 91–140. Singapore: Springer Nature Singapore.10.1007/978-981-97-6753-3_6Search in Google Scholar

Hansen, Bruce E. 1999. “The Grid Bootstrap and the Autoregressive Model.” The Review of Economics and Statistics 81 (4): 594–607. https://doi.org/10.1162/003465399558463.Search in Google Scholar

Holtemöller, Oliver, and Sushanta Mallick. 2016. “Global Food Prices and Monetary Policy in an Emerging Market Economy: The Case of India.” Journal of Asian Economics 46: 56–70. https://doi.org/10.1016/j.asieco.2016.08.005.Search in Google Scholar

Hosseinkouchack, Mehdi, and Maik H. Wolters. 2013. “Do Large Recessions Reduce Output Permanently?” Economics Letters 121 (3): 516–9. https://doi.org/10.1016/j.econlet.2013.10.012.Search in Google Scholar

Jalal, Rubia, and R. Gopinathan. 2023. “Time-Frequency Relationship Between Energy Imports, Energy Prices, Exchange Rate, and Policy Uncertainties in India: Evidence from Wavelet Quantile Correlation Approach.” Finance Research Letters 55: 103980. https://doi.org/10.1016/j.frl.2023.103980.Search in Google Scholar

John, Joice. 2015. “Has Inflation Persistence in India Changed Over Time?” The Singapore Economic Review 60 (4): 1550095. https://doi.org/10.1142/s0217590815500952.Search in Google Scholar

Khundrakpam, Jeevan. 2008. How Persistent is Indian Inflationary Process, Has it Changed? Technical report. University Library of Munich, Germany.Search in Google Scholar

Koenker, Roger. 2005. Quantile Regression, Vol. 38. Cambridge University Press.10.1017/CBO9780511754098Search in Google Scholar

Koenker, Roger, and Zhijie Xiao. 2004. “Unit Root Quantile Autoregression Inference.” Journal of the American Statistical Association 99 (467): 775–87. https://doi.org/10.1198/016214504000001114.Search in Google Scholar

Koenker, Roger, and Zhijie Xiao. 2006. “Quantile Autoregression.” Journal of the American Statistical Association 101 (475): 980–90. https://doi.org/10.1198/016214506000000672.Search in Google Scholar

Kottaridi, Constantina, Mendez-Carbajo Diego, and D. Thomakos Dimitrios. 2009. “Inflation Dynamics and the Cross-Sectional Distribution of Prices in the E.U. Periphery.” In International Trade and Economic Dynamics: Essays in Memory of Koji Shimomura, edited by Takashi Kamihigashi, and Laixun Zhao, 449–75. Berlin: Springer Berlin Heidelberg.10.1007/978-3-540-78676-4_30Search in Google Scholar

Kumar, Manmohan S., and Tatsuyoshi Okimoto. 2007. “Dynamics of Persistence in International Inflation Rates.” Journal of Money, Credit, and Banking 39 (6): 1457–79. https://doi.org/10.1111/j.1538-4616.2007.00074.x.Search in Google Scholar

Kumar, Anoop S., and Steven Raj Padakandla. 2022. “Testing the Safe-Haven Properties of Gold and Bitcoin in the Backdrop of COVID-19: A Wavelet Quantile Correlation Approach.” Finance Research Letters 47: 102707. https://doi.org/10.1016/j.frl.2022.102707.Search in Google Scholar PubMed PubMed Central

Levin, Andrew, and Jeremy Piger. 2004. Is Inflation Persistence Intrinsic in Industrial Economies? Technical report. European Central Bank.10.2139/ssrn.384584Search in Google Scholar

Leybourne, Stephen, Tae-Hwan Kim, and A. M. Robert Taylor. 2007. “Detecting Multiple Changes in Persistence.” Studies in Nonlinear Dynamics & Econometrics 11 (3). https://doi.org/10.2202/1558-3708.1370.Search in Google Scholar

Li, Guodong, Yang Li, and Chih-Ling Tsai. 2015. “Quantile Correlations and Quantile Autoregressive Modeling.” Journal of the American Statistical Association 110 (509): 246–61. https://doi.org/10.1080/01621459.2014.892007.Search in Google Scholar

Lien, Donald, and Keshab Shrestha. 2007. “An Empirical Analysis of the Relationship Between Hedge Ratio and Hedging Horizon Using Wavelet Analysis.” Journal of Futures Markets 27 (2): 127–50. https://doi.org/10.1002/fut.20248.Search in Google Scholar

Lucas, Robert E. 1980. “Two Illustrations of the Quantity Theory of Money.” The American Economic Review 70 (5): 1005–14.Search in Google Scholar

Maitra, Biswajit. 2016. “Inflation Dynamics in India: Relative Role of Structural and Monetary Factors.” Journal of Quantitative Economics 14 (2): 237–55. https://doi.org/10.1007/s40953-016-0036-5.Search in Google Scholar

Mohan, Rakesh, and Partha Ray. 2019. “Indian Monetary Policy in the Time of Inflation Targeting and Demonetization.” Asian Economic Policy Review 14 (1): 67–92. https://doi.org/10.1111/aepr.12242.Search in Google Scholar

Muduli, Silu, and Himani Shekhar. 2023. “Tail Risks of Inflation in India.” Available at SSRN 4526581.10.2139/ssrn.4526581Search in Google Scholar

Nair, Sthanu R. 2013. “Making Sense of Persistently High Inflation in India.” Economic and Political Weekly 48 (42): 13–6.Search in Google Scholar

Nair, Sthanu R., and Leena Mary Eapen. 2012. “Food Price Inflation in India (2008 to 2010): A Commodity-Wise Analysis of the Causal Factors.” Economic and Political Weekly 47 (20): 46–54.Search in Google Scholar

Oliveira, Fernando N. de, and Myrian Petrassi. 2014. “Is Inflation Persistence Over?” Revista Brasileira de Economia 68 (3): 393–422, https://doi.org/10.1590/s0034-71402014000300006.Search in Google Scholar

Özkan et al., 2024 Özkan, Oktay, Muhammad Saeed Meo, and Mehak Younus. 2024. “Unearthing the Hedge and Safe-Haven Potential of Green Investment Funds for Energy Commodities.” Energy Economics 138: 107814. https://doi.org/10.1016/j.eneco.2024.107814.Search in Google Scholar

Patel, Urjit R., S. Chinoy, G. Darbha, C. Ghate, P. Montiel, D. Mohanty, et al.. 2014. “Report of the Expert Committee to Revise and Strengthen the Monetary Policy Framework.” Reserve Bank of India, Mumbai.Search in Google Scholar

Patnaik, Anuradha. 2022. “Measuring Demand and Supply Shocks from COVID-19: An Industry-Level Analysis for India.” Margin: The Journal of Applied Economic Research 16 (1): 76–105. https://doi.org/10.1177/09738010211067392.Search in Google Scholar

Patra, Michael Debabrata, Jeevan Kumar Khundrakpam, and Asish Thomas George. 2014. “Post-Global Crisis Inflation Dynamics in India: What has Changed.” In India Policy Forum, 10, 117–203. Delhi: National Council of Applied Economic Research.Search in Google Scholar

Phiri, Andrew. 2018. “Inflation Persistence in BRICS Countries: A Quantile Autoregressive (QAR) Approach.” Business and Economic Horizons 14 (1): 97–104, https://doi.org/10.15208/beh.2018.08.Search in Google Scholar

Polanco-Martínez, J. M., J. Fernández-Macho, M. B. Neumann, and S. H. Faria. 2018. “A Pre-Crisis vs. Crisis Analysis of Peripheral EU Stock Markets by Means of Wavelet Transform and a Nonlinear Causality Test.” Physica A: Statistical Mechanics and its Applications 490: 1211–27. https://doi.org/10.1016/j.physa.2017.08.065.Search in Google Scholar

Raj, Janak, Sangita Misra, Asish Thomas George, and Joice John. 2020. Core Inflation Measures in India: An Empirical Evaluation Using CPI Data. Reserve Bank of India, Department of Economic/Policy Research.Search in Google Scholar

Roger, Scott. 2000. “Relative Prices, Inflation and Core Inflation.” IMF Working Papers 2000: 058. https://doi.org/10.5089/9781451847857.001.Search in Google Scholar

Ryczkowski, Maciej. 2021. “Money and Inflation in Inflation-Targeting Regimes – New Evidence from Time-Frequency Analysis.” Journal of Applied Economics 24 (1): 17–44. https://doi.org/10.1080/15140326.2020.1830461.Search in Google Scholar

Sethi, Chandan, and Bibhuti Ranjan Mishra. 2025. “Is Inflation Targeting Effective? Lessons from Global Financial Crisis and COVID-19 Pandemic.” International Journal of Finance & Economics 30 (3): 2327–48. https://doi.org/10.1002/ijfe.3018.Search in Google Scholar

Shahbaz, Muhammad, Arshian Sharif, Alaa M. Soliman, Zhilun Jiao, and Shawkat Hammoudeh. 2024. “Oil Prices and Geopolitical Risk: Fresh Insights Based on Granger-Causality in Quantiles Analysis.” International Journal of Finance & Economics 29 (3): 2865–81. https://doi.org/10.1002/ijfe.2806.Search in Google Scholar

Stojanovikj, Martin, and Goran Petrevski. 2021. “Macroeconomic Effects of Inflation Targeting in Emerging Market Economies.” Empirical Economics 61 (5): 2539–85. https://doi.org/10.1007/s00181-020-01987-0.Search in Google Scholar

Thornton, John. 2006. “Inflation and Inflation Uncertainty in India, 1957–2005.” Indian Economic Review 41 (1): 1–8.Search in Google Scholar

Valera, Harold Glenn A., Mark J. Holmes, and Gazi M. Hassan. 2017. “How Credible is Inflation Targeting in Asia? A Quantile Unit Root Perspective.” Economic Modelling 60: 194–210. https://doi.org/10.1016/j.econmod.2016.09.004.Search in Google Scholar

Wolters, Maik H., and Peter Tillmann. 2015. “The Changing Dynamics of US Inflation Persistence: A Quantile Regression Approach.” Studies in Nonlinear Dynamics & Econometrics 19 (2): 161–82. https://doi.org/10.1515/snde-2013-0080.Search in Google Scholar

Zhu, Huiming, Dongwei Yu, Liya Hau, Hao Wu, and Fangyu Ye. 2022. “Time-Frequency Effect of Crude Oil and Exchange Rates on Stock Markets in BRICS Countries: Evidence from Wavelet Quantile Regression Analysis.” The North American Journal of Economics and Finance 61: 101708. https://doi.org/10.1016/j.najef.2022.101708.Search in Google Scholar

Živkov et al., 2023 Živkov, Dejan, Jelena Kovačević, Biljana Stankov, and Zoran Stefanović. 2023. “Bidirectional Volatility Transmission Between Stocks and Bond in East Asia – The Quantile Estimates Based on Wavelets.” Studies in Nonlinear Dynamics & Econometrics 27 (1): 49–65. https://doi.org/10.1515/snde-2020-0113.Search in Google Scholar

Supplementary Material

This article contains supplementary material (https://doi.org/10.1515/snde-2024-0078).

Received: 2024-07-16

Accepted: 2025-07-25

Published Online: 2025-08-25

You are currently not able to access this content.

Supplementary Material Details

https://doi.org/10.1515/snde-2024-0078

Keywords for this article

inflation persistence; monetary regime; quantile regression

A Historical Perspective on India’s Inflation Persistence: A Quantile Analysis

Article

Abstract

Appendix A: Inflation and its Measurement in India

Appendix B: Rolling Regression

Appendix C: RWWQC Methodology

C.1 Wavelet Correlation (WC)

C.2 Rolling Window Wavelet Correlation (RWWC)

C.3 WQC

C.3.1 Quantile Covariance

C.3.2 Wavelet Quantile Covariance (WQC)

C.3.2.1 Wavelet Correlation at Scale s j

C.3.3 Rolling Window Wavelet Quantile Correlation (RWWQC)

Appendix D: Results Tables and Figures

D.1 Regimewise Comparison Analysis for Rolling Regression

D.2 Descriptive Statistics for CPII-IW Persistence

References

Supplementary Material

Supplementary Material

C.3.2.1 Wavelet Correlation at Scale s _j