Online Resources Accompanying
“Statistical Inference based on the Density Power Divergence”
Source Codes and Datasets (By Chapters) [List of Acronyms]
Refactored and documented using Gemini AI for better clarity.
- Compute the MDPDE under the normal model with example datasets given in Tables 2.1, 2.5 (recreating Tables 2.2 and 2.6) MATLAB Code
- Plot data and density fits (recreating Figure 2.1 ) MATLAB code
- Plot AREs of the MDPDEs under the exponential model (recreating Figure 2.2 ) MATLAB code
- Plot the IF of the MDPDE under different models, and the second order IF (IF2) and the PIF of the DPDTS for testing the mean of a normal model (recreating Figures 2.3, 2.5)
MDPDE - Normal mean MDPDE - Normal variance MDPDE - Poisson mean MDPDE - Geometric IF2 of DPDTS PIF Plotting - Compue p-values for the DPDTS under the normal model (used in Figure 2.4) Testing for variance with known mean Testing for both mean and variance jointly
- Compute the MDPDE and the p-values of the assocaited Wald-type tests under the exponential model with Leukemia data (Table 2.4) MATLAB Code (recreating Figure 2.6)
- Compute the p-values of the MDPDE based Wald-type tests for the normal mean with unknown variance (recreating Figure 2.7) Wald type test DPDTS
- Compute the MDPDE and the p-values of the assocaited tests under the Weibull model in Example 2.17 (recreating Table 2.7 and Figure 2.8) Dataset MATLAB Code
- R Functions for computation of the MDPDE, its asymptotic variance and the p-values for testing the significance of each regression coefficinet under different regression models Linear regression Logistic regression Poisson regression (These funcions can be used to recreate Tables 3.2, 3.3, 3.5, 3.6 and Figures 3.1, 3.2, 3.3)
- Calculate the data-driven optimum value of the tuning parameter τ under different regression models using the Warwick-Jones approach R function
- Datasets used in this chapter Excel file
- Plot the asymptotic contiguous power of the DPDTS under the linear regression model (recreating Figure 3.4) MATLAB code
- Compute the DPDTS and its p-vlaue to test for regression coefficients with known variances under the linear regression model (may be used to recreate Figure 3.5) MATLAB code
- Compute the DPDTS and its p-vlaue to test for regression coefficients with unknown variances under the linear regression model (may be used to recreate Figures 3.6 and 3.7) MATLAB code
- Compute the asymptotic relative efficiency (ARE) of the MDPDEs under the Poisson regression model with stochastic covariates (recreating Table 3.4) MATLAB code
- Plot the IF of the MDPDE and the second order IF and PIF of the associated Wald-type test under the Poisson regression model with stochastic covariates (recreating Figures 3.8 and 3.11) IF of MDPDE IF2 of DPDTS PIF Plotting
- Plot the IF of the MDPDE under the logistic regression model with stochastic covariates (recreating Figure 3.10) IF of MDPDE
- R Functions for computation of the MDPDE, its asymptotic variance and the p-values for testing the significance of each regression coefficinets under different regression models Linear regression Logistic regression Poisson regression (These funcions can be used to recreate Tables 4.2, 4.3, and Figure 4.5)
- Calculate the data-driven optimum value of the tuning parameter τ under different regression models using the Warwick-Jones approach R function
- Datasets used in this chapter Excel file
- Plot the IF2 and the PIF of the MDPDE based Wald-type test for testing the significance of the regression under the normal LRM with fixed design (recreating Figures 4.1 and 4.2) IF2 for contamination in fixed direction IF2 for contamination in all directions PIF Plotting
- Plot the IF of the MDPDE and the IF2 of the associated Wald-type test for testing the significance of individual regression coefficients under the Poisson regression model with fixed design (recreating Figures 4.3 and 4.4) IF for contamination in fixed direction IF for contamination in all directions IF2 for contamination in fixed direction IF2 for contamination in all directions Plotting
- Datasets used in this chapter Excel file
- Compute the MDPDE of the ACFs of a time series data (recreating Table 5.3) R Code
- Dataset used in Example 7.1 and 7.3 (recreating Figures 7.1 and 7.3) Data Plotting
- R functions for computing the MDPDE and the MPLτE under the multivariate Gaussian mixture model (recreating Table 7.1) MDPDE MPLτE
- Compute the IF of the MPLτE under the multivariate Gaussian mixture model (recreating Figure 7.3) R code
- Datasets used in this chapter Excel file
- Plot the IFs of the MDPDEs under the Exponential mean model with random right censoring (recreating Figure 8.1(a) ) MATLAB code
- Compute empirical MSEs of the MDPDEs and the empirical size/power of the associated Wald-type tests under the Exponential mean model with random right censoring (may be used to recreate Figure 8.1(b) ) MATLAB code
- Compute the MDPDEs, their asymptotic variance estimates and the p-values for the associated Wald-type tests under the Weibull model with random right censoring (may be used to recreate Tables 8.1, 8.3, 8.4 and Figures 8.3, 8.4, 8.5) MATLAB code
- Compute the MDPDEs under the PPH model with exponential baseline and random right censoring (recreating Table 8.6) MATLAB code
- Plot the AREs of the MDPDEs under the Finite Random Walk model (recreating Figure 9.1 ) MATLAB code
- Plot the AREs of the MDPDEs under the Ornstein–Uhlenbeck process (recreating Figure 9.2 ) MATLAB code
- Plot the IFs of the MDPDEs under a Poisson process model (recreating Figure 9.3 ) R code
- Plot the IFs of the MDPDEs under a drifted Brownian motion model (recreating Figure 9.4 ) R code
- Compute the MDPDEs under the geometric Brownian motion model (recreating Figures 9.5 , 9.6 and Table 9.1 ) NIFTY 500 data R code
- Datasets used in this chapter Excel file
- Compute empirical biases, MSEs and credible intervals of the ERPEs under the normal mean model with improper uniform prior (recreating Table 10.1 ) MATLAB code
- Compute the accuracy (in terms of the KLD) of the ERPDEs under the normal mean model with improper uniform prior (recreating Table 10.2 ) MATLAB code
- Compute the accuracy (in terms of the KLD) of the ERPDEs under the normal mean model with conjugate normal prior (recreating Table 10.3 ) MATLAB code
- Compute/plot the IFs of the ERPEs, the pseduo-IFs of the DPD based posterior and assocaited sensitivity measures under the normal mean model (recreating Figures 10.1, 10.2, 10.3 ) MATLAB code
- Plot the IFs of the ERPEs and the pseduo-IFs of the DPD based posterior under the normal LRM (recreating Figures 10.4, 10.5 ) MATLAB code
- Compute the ERPEs under the normal linear regression model with Jeffrey’s prior (recreating Table 10.4 ) R code
- Compute the ERPEs under the logistic regression model with normal prior (recreating Table 10.5 ) R code
More About Me
Contact
Interdisciplinary Statistical Research Unit (ISRU)
Indian Statistical Institute
R. A. Fisher Bhavan, 4th Floor.
203, B.T. Road, Kolkata – 700108, West Bengal, India.
Email: abhik.ghosh@isical.ac.in