TIME SERIES ANALYSIS AND FORECASTING OF GROSS FIXED CAPITAL INVESTMENT IN AZERBAIJAN (2010-2024)

Understanding the dynamic behavior of gross fixed capital investment is essential for evaluating short-term fluctuations and designing effective investment policies in developing and resource-dependent economies. Using annual data for Azerbaijan over the period 2010-2024, this paper applies univariate time series econometric methods to analyze the stochastic properties and short-run dynamics of capital investment. Stationarity is assessed through the Augmented Dickey-Fuller unit root test, revealing that the investment series is non-stationary in levels but becomes stationary after first differencing, indicating an integration order of one. Autoregressive and moving average specifications are employed to capture persistence and innovation-driven dynamics. The results point to a notable degree of persistence in investment behavior, while innovation effects are found to play a significant role in short-run adjustments. Diagnostic tests based on correlogram analysis and Ljung-Box statistics confirm that the selected specification adequately captures temporal dependence, yielding white-noise residuals. Overall, the findings provide a statistically consistent representation of short-run investment dynamics and offer a reliable empirical basis for short-term forecasting and policy-oriented interpretation.