![Sprocket Central
Pty Ltd
Task – 2 Data Insights
]
Data analytics approach
Vidhiya S B
Note: The dataand informationin this documentis reflectiveof a hypotheticalsituationand client.This documentis to be used for KPMG Virtual Internshippurposes only.](https://image.slidesharecdn.com/kpmgvirtualinternshiptask2-230102104948-82b2caa3-230903123707-3f6e09e9/85/kpmgvirtualinternshiptask2-230102104948-82b2caa3-pptx-1-320.jpg)
kpmgvirtualinternshiptask2-230102104948-82b2caa3.pptx
- 1. Sprocket Central Pty Ltd Task – 2 Data Insights ] Data analytics approach Vidhiya S B Note: The dataand informationin this documentis reflectiveof a hypotheticalsituationand client.This documentis to be used for KPMG Virtual Internshippurposes only.
- 2. Agenda 1. Data Exploration 2. Model Development 3. Interpretation Note: The dataand informationin this documentis reflectiveof a hypotheticalsituationand client.This documentis to be used for KPMG Virtual Internshippurposes only.
- 3. Data Exploration ▶Understand in detail the underlying data's properties for each field, such as variable distributions, whether a certain demographic is being emphasized, and the fields' applicability. ▶There are some limitations in the given datasets like some values are missing and some datatypes are different according to their value. ▶Additionally, the necessary data must be transformed such that it is in an analysis-friendly manner. Making sure the data types are appropriate and rolling data up to an aggregate level are two examples of actions that may be included in this. Or ,adding previously compiled ABS data at a geographic level to provide more variables.
- 4. ▶ Record the data's assumptions, constraints, and exclusions as well as how you might advance in the subsequent step if more time were available to address assumptions and eliminate constraints.
- 5. Model Development ▶In order to use existing data to solve the business question, we must first formulate a hypothesis connected to it.To ascertain whether or not the hypothesis is true, do statistical tests. ▶ Create calculated fields based on existing data, for example, convert the D.O.B into an age bracket. ▶ Test the performance of the model using factors like residual deviance, AIC, ROC curves, R Squared) appropriately according to the model performance, assumptions and limitations.
- 6. Interpretation and Report ▶ Presentation of the results visually. This can entail analyzing the main variables and co - efficients from a commercial standpoint. ▶This presentation gives us a general understanding of the business problem, and we use it to support our arguments with both quantitative and qualitative evidence.
- 7. Thank You