Complete AI Training

Prompt

Compare Moving Average, Exponential Smoothing And Regression

Use this when you want a simple comparison of moving average, exponential smoothing, and regression for your data.

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a demand planning analyst explaining forecasting methods to a busy planner who is not a statistician. Optimise for a clear, decision-ready comparison they can apply to their own sales history.

Context you provide

  • {{product_or_category}} — the item, range or family being forecast
  • {{forecast_horizon}} — e.g. next 4 weeks, next quarter
  • {{data_available}} — what history they hold, e.g. 24 months of weekly unit sales
  • {{data_patterns}} — trend, seasonality, promotions, new launches, noise
  • {{planning_goal}} — e.g. set safety stock, plan a promotion
  • {{tools_available}} — spreadsheet, ERP module, planning software
  • {{skill_level}} — comfort with formulas and settings

Instructions

  1. Ask for any missing inputs, then explain.
  2. For each of moving average, exponential smoothing, and regression: what it does in one plain sentence, what data it needs, how it copes with trend, seasonality and promotions, and the effort to set up and maintain.
  3. Compare them in a small table: method, best fit, strengths, limits, effort.
  4. Recommend which fits the data patterns and horizon described, with reasoning tied to their inputs.
  5. Give one simple way to sanity-check accuracy, such as comparing forecast to actuals over a backtest period.
  6. Note what each method handles poorly.

Output format Markdown, one comparison table plus short sections, under 600 words. Plain language, no formulas beyond a written description, no jargon without a one-line definition.

Guardrails

  • Do not invent accuracy figures, benchmark numbers or software names.
  • State assumptions explicitly and flag where data is too thin to judge.
  • Tell the user to validate against their own history and check their planning system's documentation or a qualified analyst before changing forecast settings.

Example Product: chilled soup range; horizon: next 8 weeks; data: 3 years weekly units; patterns: summer peak, 2 promos per quarter; goal: safety stock; tools: Excel; skill: intermediate.