AIML Rounds

QUIZ

Time series

2 questions on Time series, from 2 lessons. Pick an answer in your head, then open it.

On the roadmap for: Data Science.

  1. Question 1 of 2

    Before trusting a fancy forecasting model, what should you compare it against?

    1. A. A random forecast, to measure how noisy the data is
    2. B. Nothing: complex models always beat simple ones
    3. C. The model's own error on its training data
    4. D. A baseline like 'same as last week', on time splits
    View answer

    ✓ D. A baseline like 'same as last week', on time splits

    Many complex models fail to beat the naive forecast once tested honestly.

    Lesson: Time-Series Forecasting Basics

  2. Question 2 of 2

    What does it mean for a time series to be stationary, and how is a trend usually removed for ARIMA?

    1. A. It's always seasonal; by averaging each season
    2. B. It has no noise; by smoothing the series
    3. C. Its mean and variance stay stable; differencing
    4. D. It never changes; by deleting the trend data
    View answer

    ✓ C. Its mean and variance stay stable; differencing

    The 'I' in ARIMA is the differencing step that makes the series stationary before modelling.

    Lesson: Stationarity, ARIMA and Exponential Smoothing