Time Series Analysis Project: Analyzing UK Temperature Data with R

This project delves into time series analysis using R programming, focusing on UK temperature data from 1884 to 2019 across ten districts. The analysis begins with identifying the districts and dates with the highest and lowest average temperatures, as well as the district and year with the widest temperature range. The core of the project involves trend and seasonality analysis. Each of the ten time series is subsetted up to 2018, and the trend is estimated using linear, quadratic, and cubic regression models. The performance of these models is compared using the Akaike's Information Criterion (AIC), with appropriate plots and tables used to confirm the results. Further, the project explores averaging and sine-cosine models, and a combined linear/sine-cosine model. AIC values are used to evaluate the goodness of fit for each model, and the findings are interpreted to draw valid conclusions about the best-fit models for the temperature data. The project demonstrates practical applications of time series analysis in forecasting and statistical modeling.