Scribd Logo
Upload

Welcome to Scribd!

  • 35 views

    Time Series Analysis NMIMS

    Uploaded by

    Varsha Chotalia
    Time series analysis involves modeling and analyzing data that is collected over time. The goal is often to make forecasts about future values based on patterns in the historical data. There are different types of time series data based on frequency of collection, such as yearly, monthly, weekly, or hourly data. Time series models assume past patterns will continue into the future. Key components of time series include trends, seasonality, and irregular components. Common techniques for analyzing time series include moving averages, exponential smoothing, and ARIMA models.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd

    Time Series Analysis NMIMS

    Uploaded by

    Varsha Chotalia
    35 views17 pages
    Time series analysis involves modeling and analyzing data that is collected over time. The goal is often to make forecasts about future values based on patterns in the historical data. There are different types of time series data based on frequency of collection, such as yearly, monthly, weekly, or hourly data. Time series models assume past patterns will continue into the future. Key components of time series include trends, seasonality, and irregular components. Common techniques for analyzing time series include moving averages, exponential smoothing, and ARIMA models.

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Time series analysis involves modeling and analyzing data that is collected over time. The goal is often to make forecasts about future values based on patterns in the historical data. There are different types of time series data based on frequency of collection, such as yearly, monthly, weekly, or hourly data. Time series models assume past patterns will continue into the future. Key components of time series include trends, seasonality, and irregular components. Common techniques for analyzing time series include moving averages, exponential smoothing, and ARIMA models.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    35 views17 pages

    Time Series Analysis NMIMS

    Uploaded by

    Varsha Chotalia
    Time series analysis involves modeling and analyzing data that is collected over time. The goal is often to make forecasts about future values based on patterns in the historical data. There are different types of time series data based on frequency of collection, such as yearly, monthly, weekly, or hourly data. Time series models assume past patterns will continue into the future. Key components of time series include trends, seasonality, and irregular components. Common techniques for analyzing time series include moving averages, exponential smoothing, and ARIMA models.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    of 17

    Time

    Series
    Analysis
    ARUNANSU GORAI
    Learning Approach
    • Understand Objective –

    • Analyze Data –
    Plot, Missing values, Outliers, Structural Breaks

    • Characteristics –
    Descriptive stats, Underlying Distribution, Hypothesis testing

    • Build Models –
    Inferential methods, According to the Objective, Metrics

    • Forecasting -
    What is Time Series Analysis
    A series of data points ordered in time or data
    generated by a process over time. Time is the
    independent variable. The goal is to make a
    forecast for the future.

    Data frequency - yearly, monthly, weekly,


    hourly or even by seconds.
    Represented by (t, ) or (t, ), t = 1, 2, …., T

    Base the forecast of a variable on its own history.

    Avoid need to specify causal relationship between the dependent and independent variables.

    Combined with other independent variables, generally provide better results


    Entire theory of time series
    a) Stock Price Forecasting forecasting is based on the
    assumption that past patterns in
    the variable to forecast will
    b) Sales Forecast continue to hold in the future.

    c) Website Traffic
    Discrete and continuous time
    series

    Discrete and continuous time series


    Examples
    Examples – Original and Transformed
    Time Series Components
    Time Series Components
    Dominant Trend Component Dominant Seasonal Component

    US Annual Population Monthly Airline passengers


    Actual or
    Realized Data

    Deterministic
    Components

    Stochastic Or
    irregular
    Component
    Index Number
    Year $
    1990 1.299 Measures change over time relative to a base period
    1991 1.098 Price Index measures changes in price; e.g. Consumer Price Index (CPI)
    1992 1.087
    1993 1.067 Quantity Index measures changes in quantity; e.g. Number of cell phones produced annually
    1994 1.075
    1995 1.111
    1996 1.224
    1997 1.199
    1998 1.03
    1999 1.136  1998price   1.03 
    2000 1.484  100   100  79.3
    2001 1.42  1990price   1.299 
    2002 1.345
    2003 1.561
    2004 1.852
    2005 2.27
    2006 2.572
    Moving Next observation is the mean of all past observations upto a certain period
    Average

    Similar to moving average, but decreasing weight is assigned to each observation as we


    Exponential move into past. In other words, less importance is given to observations as we
    Smoothing move further from present.

    ARIMA Next observation is the mean of all past observations upto a certain period
    Simple Moving Average

    • Useful in smoothening time


    series to find its trend
    • Equal importance to all
    values inside MA window
    Simple Moving Average
    Exponential Smoothing and Exponential Moving Average

    • Exponential Smoothing takes all the data


    • Exponential Moving average takes only k past data points (based on its period – k)
    Exponential Smoothing

    You might also like