I will build a sales and demand forecasting model in python or excel
Data analytics, modeling optimization for business
About this Gig
Plan production, purchases and budgets with a forecast you can trust.
I'm a PhD in Economics, and my research covers forecasting of world energy prices and electric vehicle sales. I build models that capture trend, seasonality and the real dynamics of your market, and for volatile series I check trend persistence with fractal (R/S) analysis.
What you get:
- Clean historical data and a short exploratory analysis
- A forecast with 95% confidence intervals
- Accuracy metrics (MAPE) and model diagnostics
- Optimistic, base and pessimistic scenarios (Premium)
- A clear report written for managers, not data scientists
Send me your monthly or weekly data (2+ years works best) and I will tell you which package fits before you order.
Expertise:
Anomaly detection
•
Predictive analysis
Programming language:
Python
Tools:
Jupyter Notebook
•
Excel
My Portfolio
FAQ
How much historical data do you need?
Ideally 2+ years of monthly data or 6+ months of weekly data, so seasonality can be detected. With less data I can still forecast, but the intervals will be wider, and I will tell you honestly what is possible.
Which forecasting methods do you use?
It depends on your data: exponential smoothing, SARIMA, regression with seasonal terms or machine learning. I test several methods on your past data and choose the most accurate one.
Will I be able to update the forecast myself?
Yes. In the Standard and Premium packages you get the Excel model or the Python notebook with instructions, so you can rerun the forecast when new data arrives.
Can you include prices, promotions or holidays?
Yes, if you have the data. Prices, promotions, holidays and other drivers can be added to the model. Please mention them in your message before ordering.
