Research Project on "Time Series Analysis and Forecasting"
-
Time series is a sequence of observations recorded at regular time intervals.
-
Depending on the frequency of observations, a time series may typically be hourly, daily, weekly, monthly, quarterly and annual. Sometimes, we might have seconds and minute-wise time series as well, like, number of clicks and user visits every minute etc.
- Time series analysis involves understanding various aspects about the inherent nature of the series so that we are better informed to create meaningful and accurate forecasts.
- Time Series forecasting is the process of using a statistical model to predict future values of a time series based on past results.