Scribd
Upload
16 views

Lesson #8 - Time Series Analysis

Lesson #8_ Time Series Analysis

Uploaded by

chefreme420
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
16 views

Lesson #8 - Time Series Analysis

Lesson #8_ Time Series Analysis

Uploaded by

chefreme420
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

Lesson #8: Time Series Analysis

1. Introduction to Time Series Analysis

● Definition: Time series analysis involves analyzing data points collected or recorded at
specific time intervals. The goal is to identify patterns, trends, and seasonal variations to
make forecasts or understand underlying dynamics.
● Purpose:
○ To analyze and model temporal data.
○ To forecast future values based on historical data.
○ To identify seasonal patterns and long-term trends.

2. Key Concepts in Time Series Analysis

● 1. Time Series Components:


○ Trend:
■ The long-term direction or movement in the data over time.
■ Example: A steady increase in sales over several years.
○ Seasonality:
■ Regular, periodic fluctuations that occur at specific intervals (e.g., daily,
monthly, yearly).
■ Example: Increased ice cream sales during the summer months.
○ Cyclical Patterns:
■ Fluctuations that occur over longer periods, typically linked to economic
or business cycles.
■ Example: Business growth and decline phases over several years.
○ Irregular (Random) Component:
■ Unpredictable variations or noise in the data that do not follow any
pattern.
■ Example: Sudden spikes in sales due to unexpected events.
● 2. Stationarity:
○ A time series is stationary if its statistical properties, such as mean, variance, and
autocorrelation, are constant over time.
○ Importance: Most time series models assume stationarity; non-stationary series
need to be transformed (e.g., differencing) before modeling.
○ Tests for Stationarity: Augmented Dickey-Fuller (ADF) test, KPSS test.
● 3. Autocorrelation:
○ Measures the correlation between a time series and a lagged version of itself.
○ Autocorrelation Function (ACF): Plots the autocorrelation of a time series for
different lags.
○ Partial Autocorrelation Function (PACF): Measures the correlation between a
time series and its lags, accounting for the contributions of intermediate lags.

3. Time Series Models

● 1.1 Moving Average (MA) Model:


○ Definition: Models the time series as a linear combination of past errors
(residuals).
○ Equation: Yt=μ+ϵt+θ1ϵt−1+⋯+θqϵt−qY_t = \mu + \epsilon_t + \
theta_1\epsilon_{t-1} + \dots + \theta_q\epsilon_{t-q}Yt=μ+ϵt+θ1
ϵt−1+⋯+θqϵt−q
■ μ\muμ = Mean of the series.
■ ϵt\epsilon_tϵt = Error term (random noise).
■ θ1,θ2,…,θq\theta_1, \theta_2, \dots, \theta_qθ1,θ2,…,θq = Coefficients
for the past errors.

You might also like