Scribd
Upload
3 views2 pages

Time Series Transformation Assignment

The assignment involves exploring power transformations on a dataset of monthly sales to stabilize variance. It requires applying specific transformations, calculating variances, and visualizing the results to determine which transformation best stabilizes variance. Additionally, it asks for an explanation of the popularity of log transformations and a recommendation for the best transformation for a dataset representing monthly airline passengers.

Uploaded by

syedazadarhussain5121472
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
3 views2 pages

Time Series Transformation Assignment

The assignment involves exploring power transformations on a dataset of monthly sales to stabilize variance. It requires applying specific transformations, calculating variances, and visualizing the results to determine which transformation best stabilizes variance. Additionally, it asks for an explanation of the popularity of log transformations and a recommendation for the best transformation for a dataset representing monthly airline passengers.

Uploaded by

syedazadarhussain5121472
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

Assignment: Exploring Data

Transformation in Time Series


Instructions: Use the data below or your own small dataset to explore the effect of power
transformations on variance stabilization. Show all calculations and plots where necessary.

Part 1: Apply Power Transformations


You are given the following dataset of monthly sales (units):

Month Sales

1 100

2 120

3 150

4 200

5 300

6 500

Compute the transformed values for the following λ values using the formula:

y(λ) = (y^λ - 1)/λ , λ ≠ 0


ln(y) , λ = 0

λ values to use: 1 (no transformation), 0.5 (square root), 0 (log transformation)

Create a table of transformed values alongside the original data for each λ.

Part 2: Compare Variance


Compute the variance of the original series and for each transformed series. Discuss which λ
value stabilizes the variance best.

Part 3: Visual Analysis


Plot the original series and each transformed series (one plot per λ). Comment on how the
transformations affect the spread of data and visibility of trend.

Part 4: Application and Interpretation


Explain why log transformation (λ=0) is popular for time series with increasing variance. If
the dataset represents monthly airline passengers, which transformation would you use
and why?

Submission Requirements
Submit a PDF or Jupyter Notebook showing:

- Tables of transformed values

- Variance calculations

- Plots for visual comparison

- Short discussion of results (100–150 words)

You might also like