Scribd
Upload
9 views9 pages

Pad Ibm Assignment 1

Uploaded by

saiprudhvibodempudi11
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
9 views9 pages

Pad Ibm Assignment 1

Uploaded by

saiprudhvibodempudi11
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

Assignment-1

Describe and elucidate the practical application of temporal data using time series. Explain
the different components of the time series.
Submit a python program for a use case of your choice using the Python speech and audio
analysis tools such as Speech Recognition or PyAudio.

TIME SERIES FORECASTING STEPS


Step 1
Clearly define the problem and the intended goals. Then you can decide how to tackle the
project, including which software to use
Step 2
Now your data has been prepared you can start to analyse it. Time series can be decomposed
into several components, including trend and seasonality
Step 3
Analyse your collected ts data. Time series can be decomposed into several components,
including trend and seasonality.
Step 4
Build your time series model by combining the trend and seasonality to create a forecast.
Step 5
Deploy and make the forecasts live. Using a dashboard or web application.
Step 6
Monitor the forecast performances.

Stationarity
This is a very important concept in time series analysis.
To apply time series model, it is important for the time series to be stationary
The variance and mean of the data should be constant and covariance is not a function of
time.
How to check the stationarity of data.
In time series forecast time is the important factor y =f(t)
Variable of t is affected by time.
TIME SERIES DATA COLLECTION
You need to collect ts data in the least available aggregation of data is possible.
Time series components has two classifications
Systematic and non-systematic parts.
The systematic parts contain level(average), trend (Upward or downward) and seasonal
patterns (Cycles)
Non systematic part has no patterns and has noise.

Time series data components


Level trend and seasonality are commonly comsidered to be addictive or multiplicative.
Time series with additive components can be written as yt=level + trend + seasonality+ noise
Time series with multiplicative components can be written as yt=level*trend*seasonality*
noise

Reading Time Series Objects (Air Passengers)

Reading Time Series Objects (India GDP Data)


Exploring Types of Time Series Data: Univariate
Exploring Types of Time Series Data: Multivariate
Time Series Components: Trends

Time Series Components: Seasonality


Time Series Decomposition: Additive Model

Decompose the time series.


Time Series Decomposition: Multiplicative Model
Visualization of Seasonality

You might also like