SIG$742$: Modern Data Science Assignment $2$ Trimester $2,2024,$ Deakin University

Need solution of below problem statement

Sales Prediction There are $3$ questions in this part, totaling $40$ marks. Each question is worth $10$ marks. Additionally, the quality of your explanation in both the report and video will collectively be worth $10$ marks. You are required to use Google Colab to finish all the coding in the code block cell, and provide sufficient coding comments, and also save the result of running as well. In this part, we will focus only on two columns revenue with transaction$\_$date to form the revenue time series based on transaction$\_$date. We will use the dataframe df from question $1\mathrm{.}2($without any filtering on transactions$)$ to finish below sub$-$tasks:

Question $2\mathrm{.}1$ You are required to explore the revenue time series. There are some days not available in the revenue time series such as $2019-01-01.$ Please add those days into the revenue time series with default revenue value with the mean value of the revenue in the whole data $($without any filtering on transactions$).$ After that, decompose the revenue time series with addictive mode and analyses on the results to find if there is any seasonality pattern $($you could leverage the M$05$A material from lab session with default setting in seasonal$\_$decompose function$).$