kacperkerszen / GDP_time_series

GDP time series analysis using R for understanding and forecasting the behavior of GDP and its components over time, aiding in economic analysis and decision-making.

Time series analysis of Poland's GDP using R

Time-series analysis for understanding and forecasting the behavior of GDP and its components over time, aiding in economic analysis and decision-making.

Overal Project Description

• The code covers a comprehensive workflow for time series analysis and modeling, including data preprocessing, exploratory analysis, stationarity testing, ARIMA modeling, and model diagnostics.
• It utilizes various R packages such as timeSeries, tseries, forecast, zoo, xts, Quandl, quantmod, and urca for different tasks.
• Techniques such as ACF, PACF, differencing, logarithmic transformation, and statistical tests are employed to analyze and model the time series data.

Data Preprocessing

• Libraries necessary for data handling, visualization, and modeling are installed and loaded.
• Data is read from an Excel file (R1.xlsx) using the read_excel function from the readxl package.
• The data is viewed and extracted into relevant variables, such as quarterly GDP (Y) and consumption (C).
• Quarterly increments of GDP and consumption are calculated.

Exploratory Data Analysis

• Simple time series plots are generated to visualize the GDP level and its growth rate.
• Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF) plots are created to analyze the autocorrelation structure of the time series.

Stationarity Analysis

• The concept of stationarity is introduced with theoretical examples of white noise and random walk processes.
• The logarithm of GDP (logY) and its increments (dlogY) are analyzed and plotted.
• Stationarity tests including Augmented Dickey-Fuller (ADF), Phillips-Perron (PP), and Kwiatkowski-Phillips-Schmidt-Shin (KPSS) tests are performed on both the original and differenced series.

ARIMA Modelling

• Automatic ARIMA model selection is performed on both logY and dlogY series using the auto.arima function.
• Manual ARIMA model fitting is demonstrated with the arima function.
• Diagnostic checks on the ARIMA models are conducted, including plotting the residuals, autocorrelation of residuals, and normality tests.
• Model-based forecasting is carried out for future time periods using the forecast function.

Alternative ARIMA Model Estimation

• Different ARIMA model estimation methods (CSS-ML, CSS, ML) are demonstrated.
• Coefficients of the ARIMA models are summarized.