This repository contains a paper on application of seasonal ARIMA model for forecasting GDP of United Kingdom. To preview the paper, click on the file paper.pdf. PDF was generated using R Markdown. For clean R code click on the file code.R.
- initial analysis using descriptive statistics
- determining time series stationarity with ADF and HEGY tests
- estimating seasonal ARIMA models and chosing the best one based on AIC
- interpretation of error measures
- CUSUM and Chow stability tests
library(readxl) # for xlsx loading
library(tidyverse) # vide range analytical tools
library(zoo) # conversion to time series
library(urca) # ADF test with restrictions
library(car) # Linear restrictions test
library(moments) # additional descriptive statistics
library(forecast) # forecasting
library(strucchange)# CUSUM tests