Two forecasts. Zero historical data. Here’s how I built them.
📰 Medium · Data Science
Learn how to build forecasts without historical data, a crucial skill for data scientists and analysts
Action Steps
- Identify the problem domain and understand the requirements
- Determine the type of forecast needed (e.g. time series, regression)
- Explore alternative data sources (e.g. similar products, markets, or industries)
- Apply machine learning techniques (e.g. transfer learning, meta-learning) to build the forecast model
- Evaluate and refine the model using metrics (e.g. mean absolute error, mean squared error)
Who Needs to Know This
Data scientists, analysts, and product managers can benefit from this skill to make informed decisions in the absence of historical data
Key Insight
💡 Forecasts can be built without historical data by leveraging alternative data sources and machine learning techniques
Share This
📊 No historical data? No problem! Learn how to build forecasts from scratch #datascience #forecasting
Key Takeaways
Learn how to build forecasts without historical data, a crucial skill for data scientists and analysts
Full Article
The most uncomfortable question you can ask a data scientist is: can you forecast this? — when “this” has never existed before. Continue reading on Medium »
DeepCamp AI