Subtitle: Understanding Trends and Patterns Over Time Your Name: Hazair Abdullah (22012120-036) Hussain Imran (22812120-001) Hasan Saeed (22012120-040) Introduction to Time Analysis Definition: Examination of data points collected or recorded at specific time intervals.Importance: Identifies trends, seasonal patterns, and cyclic behaviors. Applications: Economics, finance, weather forecasting, healthcare. Types of Time Series Data Univariate vs. Multivariate: Single variable versus multiple variables. Regular vs. Irregular Intervals: Data recorded at consistent versus inconsistent intervals. Components of Time Series Trend: Long-term movement in data. Seasonality: Regular, repeating patterns or cycles. Cyclic Patterns: Long-term fluctuations due to economic cycles. Irregular/Noise: Random variation or "noise.” Time Series Decomposition Additive Model: ( Y(t) = T(t) + S(t) + E(t) ) Multiplicative Model: ( Y(t) = T(t) \times S(t) \times E(t) ) Explanation: Breaking down data into trend (T), seasonal (S), and error (E) components. Stationarity in Time Series Definition: Statistical properties (mean, variance) do not change over time Importance: Many time series models assume stationarity.Methods to Achieve Stationarity: Differencing, detrending, and transformation. Autocorrelation and Partial Autocorrelation Autocorrelation Function (ACF): Correlation between observations at different lags.Partial Autocorrelation Function (PACF): Correlation between observations at different lags, controlling for shorter lags. Time Series Models ARIMA Model: AutoRegressive Integrated Moving Average. AR (AutoRegressive): Relationship between an observation and a number of lagged observations. I (Integrated): Differencing of raw observations to make the series stationary. MA (Moving Average): Relationship between an observation and a residual error from a moving average model.SARIMA Model: Seasonal ARIMA, includes seasonal components. Model Evaluation Criteria: AIC (Akaike Information Criterion), BIC (Bayesian Information Criterion), RMSE (Root Mean Square Error). Cross-Validation: Splitting data into training and testing sets to evaluate model performance. Forecasting Definition: Predicting future values based on historical data.Techniques: Exponential Smoothing, Holt-Winters, ARIMA. Example: Sample forecast plot. Thank you