Sales Forecasting Data Science Project

I have dataset and want to create capstone project for my submission. The data is CRM data and we would like to understand following:


1.    To study impact of Covid on Sales Q1 FY 19 to Q1 FY 20

2.    Analysis for FY 19

3.    Forecast for FY 20

4.    Launch of New Services and Product – Customer Segmentation


Need help for full project report. I will share source data files once we freeze. Attached is capstone project details required. 

Get Help With a similar task to - Sales Forecasting Data Science Project

Login to view and/or buy answers.. or post an answer
Additional Instructions:

Business Analytics for Senior Managers: Capstone Project Guidelines This document is intended to provide a guideline to students to complete their Capstone Project for the course Business Analytics for Senior Managers. For completing the project having a relevant dataset is a key requirement. We recommend participants to arrange for data from the companies, government or public institutions, not for profit organisations, etc. they work for or are associated with in any capacity. This will help them to get the experience of doing a live project as well as adding value to their companies or clients. In case it is not possible for a participant to arrange for data from such primary sources, data sets publicly available like those in Kaggle (https://www.kaggle.com/) may be used. The broad sections and the content expected in the project is given below. The main section of the document should talk about the problem, strategy to solve, execution methods, and the key outcomes. The intermediate detailed steps can come as annexure to the main project. Codes written, screenshots from tools should also come in annexure. All annexures should be numbered. Any reference to the annexure on the main body of the project should be against the annexure number. The focus of the capstone project is to identify the right business case for application of analytics and the benefits the project is expected to bring to business. Business value of the project will be the key driver in ascertaining the credit to the project rather than factors like the complexity of model designed, the technology used to implement the solution, the difficulty of the code written, etc. Background In this section of the project provide a background to the project you plan to do. The background may include information about the company (in case you are doing the project on a company) or the organisation or country. It can also talk about the world issue (pollution, COVID-19, etc.) on which you are anchoring your project. Problem Definition and Business Case In this section you will define the business problem you are trying to address through analytics e.g. when Covid-19 new cases can be expected to come down below a certain number, predicting if a customer will churn for a telecom company, where will an ecommerce company set up its hubs, etc. You can have more than one problem defined. But all the problems you define here need to be given a reasonable solution. Thus, you may decide to be prudent in selection of the nature and the number of problems. You should select those problems for which you have relevant data to work on and have a broad idea of the approach to take to solve the problem. https://www.kaggle.com/ You would also need to provide the business case for solving the problem i.e. the return on investment for the project. You can use some of the frameworks discussed in class. You may not have all the necessary data or resources to build the complete business case and calculate the financial metrics. In such cases providing the approach to business case will suffice. Technology Stack Selection In this section you would need to select the stack of technologies that you would need to design, develop and implement the solution to the business problem you have identified. The technology stack can be as simple as just Excel. You may consider more complex software, platforms, packages and stacks e.g. R, Python, Knime, RapidMiner, Jupyter Notebook, AWS Sagemaker, Hadoop cluster and Mapreduce, Spark, etc. To re-emphasize, credit given to the project is not based on the complexity of the technology landscape chosen but the business value generated. A simple spreadsheet model can provide significant business value similar to a complex technology. Data Collection and Pre-processing In this section talk about the method that you have adopted to collect the data. For example, was the data from a backend ERP system or from an excel form. Was the data structured or unstructured? What was the volume of data? What was the strategy taken to extract the data? What was the effort required to extract the data? How did you estimate it? What kind of resources were required to extract the data and how did you arrange for the resources? Once you have collected the data, talk about the quality of data and the challenges faced, if any. What action did you take on the missing data and what were the basis of these actions? What pre-processing was necessary for the data. The pre-processing and model selection steps can be iterative as some of the pre- processing (e.g. whether input features require normalisation or not) depends of the model that has been selected. Use tools and functionalities of your choice and comfort (Excel, Knime, Python etc.) to pre- process the data. Model Selection In this section talk about the class of analytics (classification or clustering; supervised, unsupervised, deep learning, reinforced learning, etc.) that you are plan to consider for applying to the identified problem. Deep dive into the model the shortlist of models that you plan to implement (logistic regression, decision tree, neural network, support vector, k- means, association rule, etc.). Define the metrics that you will consider to measure the performance of the model. Provide adequate statistical basis for your selection. Mention relevant feature engineering (e.g. principal component analysis) that might be required on the data set to get the best result out of the model. In case of supervised learning you can also talk about the method that you are adopting to train the model and how you plan to divide the dataset into training, validation and testing data sets. Execution and Model Tuning In this section talk about the execution of the model. You can provide the source code in the annexure. You can also provide the screenshots of the intermediate steps and output in the annexure. Explain how you did the tuning of the models. How do you compare the performance of the models? Which model would you select to productionise? Data Visualisation and Story Telling Use appropriate tools to visualise the data. Tell a story around the data. Give screenshots of the visualised data in the annexure Management Action Based on the analytics done so far recommend suitable management action to solve the problem you started with. Conclusion Conclude with the learnings from the project.

Related Questions

Similar orders to Sales Forecasting Data Science Project
16
Views
0
Answers
Python case study
## Download the iPython Notebook and data from Canvas: 0. Make sure you have Anaconda installed and working (this is useful to follow the class demonstrations anyway) 1. We made a zip file containing both the data and the iPython Notebook file avail...
35
Views
0
Answers
python
Write a program that allows the user to enter any number of decimal numbers. Then the program should print the 3 highest numbers sorted by the highest number first. When entering the numbers, you must be able to type 'x' to end the entry...
21
Views
0
Answers
Basic Excel functions (30 min)
Intermediate Excel knowledge and functions...
81
Views
0
Answers
Hematology exam uses proctorU
I have hematology exam uses proctorU...