# Compare absolute and squared forecast errors

Illustrative planning brief; no automatic product import.

Synthetic planning case: Three duration forecasts are two days each. The observed toy outcomes are three, three and six days, giving errors one, one and four.

## Decision

The invented sample has a two-day MAE and approximately 2.45-day RMSE. The squared-error measure gives the four-day miss more influence; neither number selects a universal objective.

## Owned work

- Match units and outcomes
  - Owner role: Evaluation owner
  - Acceptance evidence: All three errors describe the same duration target in days.
- Calculate both declared summaries
  - Owner role: Reviewer
  - Acceptance evidence: Absolute and squared contributions retain the large case-C miss.
- Choose the decision objective
  - Owner role: Planning lead
  - Acceptance evidence: The preferred loss rule is discussed in relation to actual consequences, not selected because its number looks lower.

## Workflow

1. Match units and outcomes. Check: All three errors describe the same duration target in days.
2. Calculate both declared summaries. Check: Absolute and squared contributions retain the large case-C miss.
3. Choose the decision objective. Check: The preferred loss rule is discussed in relation to actual consequences, not selected because its number looks lower.

## Judgment

This tiny synthetic set is not a model comparison study. The rule and affected decision must be justified separately.


## Filled manual planning note

The invented sample has a two-day MAE and approximately 2.45-day RMSE. The squared-error measure gives the four-day miss more influence; neither number selects a universal objective. MAE=(1+1+4)/3=2 days. RMSE=√((1+1+16)/3)=√6≈2.45 days. Units return to days after taking the square root. This tiny synthetic set is not a model comparison study. The rule and affected decision must be justified separately.


## Before / after

Before: Absolute contributions 1,1 and 4 give MAE two days.

After: Squared contributions 1,1 and 16 give RMSE√6 days; the objective changes the influence of case C.


## Workflow questions

### Can I directly compare an MAE number with RMSE to pick a model?

They summarize different loss rules; compare models under the same declared objective and compatible cases.

### Why is the squared-error column in days squared?

Squaring a duration changes units; RMSE takes a square root and returns to days.

## Product connection

Use the owned checks and downloaded brief to discuss this planning decision alongside your TeamBoostAI tasks. Confirm available fields, roles and account features separately. The example is manual; it does not calculate live analytics, create work or run an experiment in the product.

Confirm account availability before adopting this manual outline.

## Original worked case

Synthetic records, manual planning only. No account import or live analytics.

### Inspect the invented case records

Case | Forecast days | Actual days | Absolute error | Squared error
--- | --- | --- | --- | ---
A | 2 | 3 | 1 | 1
B | 2 | 3 | 1 | 1
C | 2 | 6 | 4 | 16

### Reasoning

MAE=(1+1+4)/3=2 days. RMSE=√((1+1+16)/3)=√6≈2.45 days. Units return to days after taking the square root.

### Bounded result

The invented sample has a two-day MAE and approximately 2.45-day RMSE. The squared-error measure gives the four-day miss more influence; neither number selects a universal objective.

### Distinct decision

The artifact exposes how one large miss changes two explicitly defined error summaries.

### Limits

This tiny synthetic set is not a model comparison study. The rule and affected decision must be justified separately.

### Definitions and method context

- Forecasting: Principles and Practice — accuracy — https://otexts.com/fpp3/accuracy.html — Genuine held-out forecast and point-error evaluation context. Binary scoring examples use their own explicit toy definitions; no real task model is validated.