The index of economic freedom dataset (2026 version) includes 12 variables for each country which fall into 4 pillars:Pillar#1 Rule of law- property rights, judicial effectiveness, and government integrity.Pillar#2 Government Size- tax burden, government spending, and fiscal health.Pillar#3 Regulatory Efficiency- Business freedom, labour freedom, and monetary freedom.Pillar#4 Open Markets- trade freedom, investment freedom, and financial freedom. One would refer to these 12 variables as the pillar variables. All these variables are index scores from 0 to 100. They are combined into an overall score and the name of the country and region of the world (Pacific, Europe, etc) are also given. See https://www.heritage.org/index/pages/about for further details. Task- Linear Models.Make sure you work through the checkpoint #2 Example and Exercises before you attempt this task. Include any R code and output directly in your report (not as an appendix). Where possible you must use R code from the R tidytidyverse (as used in lab worksheets for this module). Eg use read_csv() nnot read.csv(). Dataset for task 2: economic_freedom_2026_indieconomic_freedom_2026_individual.csv. you wish to use the 12 pillar variables to predict overall score using linear models(regression) in R.we are primarily concerned with critically assessing any linear models proposed and with model selection (which predictors to include in any final linear models recommended).use the R package “olsrr” (where appropriate) to help carry out this task (see https://cran.r-protect.org/web/packages/olsrr/vignettes/variable selection.html). Throughout this task, please take every opportunity to investigate the effect of region in your plots. 1) Briefly explain the acronym “AIC” used to compare models, why it is needed and how it is defined. Make sure you cite any sources used. 2) consider the following linear models.compare these models and summarise your results in a compact table. Include any R code and output directly in your report. You must clearly (i) justify, (ii) specify, (iii)fit, (iv) interpret and (v) critically evaluate all of the linear models that you suggest. Briefly discuss and compare what you notice about the predictors included in each model. Model #1: The linear model using only fiscal health and trade freedom to predict overall score? JUSTIFY your answer. Model #2: The best linear model to predict overall score that uses at most four of the 12 pillar variables as predictors. Model #3: The best linear model to predict overall score that uses variables from at most two of the pillars . Model #4: The best two predictor linear model to predict overall score for each region separately. How do these predictors relate to the pillars and which of these would you recommend as the best model? 3) considering the diagnostic plots produced by autopilot () corresponding to at most two (carefully selected) linear models from part (2). Discuss whether any of the countries in the dataset need further investigation. Fully justify any recommendations you make (be specific). Including any criteria you have applied.
Leave a Reply
You must be logged in to post a comment.