WEEK 4 ARTICLE THE COMPLETE GUIDE TO TIME SERIES MODEL.

WEEK 4 ARTICLE
THE COMPLETE GUIDE TO TIME SERIES MODELS TIME DATA MODEL – A COMPLETE GUIDE
WHAT IS TIME SERIES MODEL?
Time series model is a set of data points ordered in time, where time is the independent variable used to analyze and forecast the future. It comprises of a process and mathematical tools used for analyzing time series data to learn about what happened, when, why it happened and what is most likely to happen.
CHARACTERISTICS OF TIME SERIES MODELS
It helps to know their three main characteristics: autocorrelation, seasonality and stationarity.

1. Auto-correlation
   This is a measure of the target variable at different points in time are linearly related to one another as a function of time difference.

2. Seasonality
   This is a pattern that reoccurs in a data at fixed intervals either as an increasing or decreasing trend (as variation or fluctuations specific to a particular time frame)

3. Stationarity
   This is a concept that describes the time series whose statistical properties (mean, variance) are constant over time and its covariance is independent of time.

Time Series Data Visualization Methods
Data visualization is conducted with special tools that offer users a variety of visualization kinds and format these include:
• Time series graph
Time series graphs for illustrating data points at a temporal scale where each point corresponds to both time and the unit of measurement.
• Real time graph
Time series data is displayed using real time graphs known as data streaming charts that automatically refreshes every seconds or when a new data point is received from the server.

DATA MODELS FOR TIME SERIES

1. Autoregressive (AR) Model AR model is a representation of a form of random process used to describe data reflecting time-varying processes example changes in weather, economics.
2. Integrated Models Integrated models are made up of a series of random walk components a summation of weakly steady components.
3. Moving-average (MA) models Univariate time series are modeled using moving-average models. The output variable in MA models is linearly dependent on the current and various historical values of an imperfectly predicted (stochastic) factor.
4. Autoregressive moving average (ARMA) models ARMA models combine the AR and MA classes, with the AR part regressing the variable on its own historical values and the MA part modeling the error term as a linear mixture of error terms happening concurrently and at different times in the past. They are applied in analytics for forecasting future values in a series.
5. Autoregressive Integrated Moving Range (ARIMA) Model This is used when data show evidence of non-stationarity, where an initial differencing step, corresponding to the integrated part of the model applied to eliminate the mean function's non-stationarity.
6. Autoregressive Fractionally Integrated Moving Average (ARFIMA) ARFIMA models are used to simulate so-called long memory time series deviations from the long-run mean dissipate more slowly than exponential decay.

Limitations of Time Series Model

1. Generalizations from a single or small sample of subjects should be done carefully example finding the time a customer example finding the time a customer is most likely running requires analyzing run frequencies of many customers
2. Predicting future values may be impossible if the data was not properly prepared and irregularities in data may arise in the future.
3. It does not allow exact predictions but only probability distributions of specific values e.g weather forecast can be in accurate.
4. Dealing with missing values Time-series data frequently contains missing or partial values, which can impair analysis and modeling accuracy.
5. Overcoming noise in time-series data Noise in time series data refers to random fluctuations or anomalies that can conceal underlying patterns and trends.