Part 1: Basic Introduction About Time-Series Forecasting
Time Series Forecasting is the use of the statistical model to predict future value based on past observation/results. Use any variables tracked and collected over time. For example, annual population, daily stock prices, daily-weekly-quarterly sales, etc.
Time Series Data Characteristics & Objectives:
- The data completely depend on data and time. Hence, if the order of date or time change leads to a change in the meaning of data.
- Data collection must be done sequentially and at an equal interval of time.
- Once data collection is done properly, the next step is to identify patterns with help of a sequence of observations and forecast/predict the future values in Time series.
Time Series Components
1. Time Series Plot
In Time-Series plot always, X-axis indicates Date/Time and Y-axis indicates Target Variable. For example, Want to predict monthly sales of the shop. Then X-axis: Days and Y-axis: Per day sales value.
2. Types of Trend in Time-Series plot
There are 3 types of trend can be observed in Time-Series graph. Also, each observed trend in Time-Series called Trend Cycle.
- Uptrend: The trend having HH (Higher High) & HL (Higher Low).
- Downtrend: The trend having LH(Lower High) & LL (Lower Low).
- Horizontal or Sideway Trend: The trend moving with same or no variation in time series and moving horizontally.
- Time-Series exhibits a repeated pattern at a fixed interval of time. For example, Sales of Air-Conditioner increase during summers and decrease during winters.
- Seasonality trend helps business to manage their inventories, staffs and other business-related decision.
3. Cyclical Pattern
- Time-Series has no fixed period for rising/fall pattern.
- Also, there is no repeated pattern within each year i.e no seasonality observed.
- The magnitude of trend cycles changes more frequently, hence much harder to predict the future trend.
Types of Time-Series Predictive/Forecasting Models
- ETS — Error Trend Seasonality Model
ETS models are non-stationary. ETS models are considered as exponential smoothing and then state-space. ETS model describes how unobserved components of the data (error, trend, and seasonality) change over time.
In the ETS model giving more weight to most recent observations or values with weights gradually gets smaller as observations get older.
There are 4 types of ETS methods:
- Simple Exponential Smoothing Method (SES)
- Holt’s Linear Trend Method (HLT)
- Exponential Trend Method (ET)
- Holt’s-Winter Seasonal Method (HWS)
2. ARIMA — Auto-Regressive Integrated Moving Average Model
ARIMA models are stationary. ARIMA models focus on autocorrelation in data and forecast future trends.
There are two types of ARIMA models i.e. non-seasonal and seasonal.
Where: p = Auto regressive value,
d = Number of transformation done to make data stationery,
m = Moving Average
ARIMA (p,d,m) (P,D,M) m
p: Autoregressive value,
d: Number of transformation done to make data stationery,
m: Moving Average
P, D, M: Autoregressive, number of transformation done to make data stationery, Moving Average of Seasonal Data.
m: Number of periods in each season
In this blog, I try to summarise the Time-Series its characteristics, types and about forecasting model. In upcoming blogs, I will explain each Time-Series Forecasting models (ETS & ARIMA) in details with Implementation using python.