Support me with an R code
Budget: €8 – €30 EUR
So my goal is to replicate the following paper: "Inflation at risk in advanced and emerging market
economies by Ryan Banerjee, Juan Contreras, Aaron Mehrotra and Fabrizio Zampolli" by using my own data set. I also have panel data for different countries but my dependent variable is foreign exchange rate.
I want to run the code in R and have done the quantile regressions. Now, my problem is as in the post described here: https://stats.stackexchange.com/questions/478354/quantile-matching-using-the-skewed-t-distribution-from-azzalini-capitanio-200
I try to replicate some findings from a paper (page 11/12 of https://www.newyorkfed.org/medialibrary/media/research/staff_reports/sr914.pdf). In this paper they estimated some quantiles using quantile regressions. I did this step, so I have the 5%, 25%, 75% and 95% quantiles for multiple periods and variables. Then, they used these estimated quantiles to fit a conditional probability distribution for every period and variable. The probability distribution used is the four-parameter skew t-distribution of Azzalini & Capitanio (2003). I found the R package, which generates this kind of distribution (https://cran.r-project.org/web/packages/sn/sn.pdf).
So what I need is:
To get from the Empirical CDF to the T-skew Parameters - I want to estimate the t-skew distributions. For that the parameters of interest are derived by minimizing the distance between the empirical quantiles and the quantiles of a t-skew. Once the optimal t-skew parameters have been estimated from the conditional quantiles, I need to derive the fitted t skew CDF and probability density function (PDF). And do the CDF and PDF plots just like in the Banerjee et al (2020) paper.
economies by Ryan Banerjee, Juan Contreras, Aaron Mehrotra and Fabrizio Zampolli" by using my own data set. I also have panel data for different countries but my dependent variable is foreign exchange rate.
I want to run the code in R and have done the quantile regressions. Now, my problem is as in the post described here: https://stats.stackexchange.com/questions/478354/quantile-matching-using-the-skewed-t-distribution-from-azzalini-capitanio-200
I try to replicate some findings from a paper (page 11/12 of https://www.newyorkfed.org/medialibrary/media/research/staff_reports/sr914.pdf). In this paper they estimated some quantiles using quantile regressions. I did this step, so I have the 5%, 25%, 75% and 95% quantiles for multiple periods and variables. Then, they used these estimated quantiles to fit a conditional probability distribution for every period and variable. The probability distribution used is the four-parameter skew t-distribution of Azzalini & Capitanio (2003). I found the R package, which generates this kind of distribution (https://cran.r-project.org/web/packages/sn/sn.pdf).
So what I need is:
To get from the Empirical CDF to the T-skew Parameters - I want to estimate the t-skew distributions. For that the parameters of interest are derived by minimizing the distance between the empirical quantiles and the quantiles of a t-skew. Once the optimal t-skew parameters have been estimated from the conditional quantiles, I need to derive the fitted t skew CDF and probability density function (PDF). And do the CDF and PDF plots just like in the Banerjee et al (2020) paper.