Time series components (trend, seasonality) in ML Python - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is the trend component in a time series?

The trend is the long-term increase or decrease in the data over time. It shows the overall direction, like a steady rise or fall.

beginner

Define seasonality in a time series.

Seasonality is a repeating pattern or cycle in the data that happens at regular intervals, like daily, weekly, or yearly.

intermediate

How does seasonality differ from trend in time series data?

Trend shows the overall direction over a long time, while seasonality shows repeating patterns within shorter fixed periods.

intermediate

Why is it important to separate trend and seasonality in time series analysis?

Separating them helps us understand the true underlying patterns and make better predictions by modeling each part correctly.

beginner

Give a real-life example of seasonality in time series data.

Sales of ice cream often rise every summer and fall in winter, showing a yearly seasonal pattern.

What does the trend component in a time series represent?

A Sudden spikes in data

B Random noise in data

C Repeating patterns at fixed intervals

D Long-term increase or decrease in data

Trend shows the overall direction of data over a long period.

Which of the following is an example of seasonality?

A Stock price steadily rising over years

B Daily temperature rising and falling every day

C Weekly sales increasing every weekend

D Random fluctuations in sensor readings

Weekly sales increasing every weekend is a repeating pattern at fixed intervals, which is seasonality.

Why do we separate trend and seasonality in time series analysis?

A To better understand and predict data patterns

B To remove all data points

C To ignore seasonal effects

D To increase noise in data

Separating trend and seasonality helps us model and predict data more accurately.

Which component is NOT part of typical time series decomposition?

A Clustering

B Trend

C Noise

D Seasonality

Clustering is a machine learning technique, not a time series component.

If sales rise every December, this pattern is called:

A Trend

B Seasonality

C Noise

D Outlier

A repeating yearly pattern like sales rising every December is seasonality.

Explain the difference between trend and seasonality in time series data.

Think about how data moves over years versus repeating cycles like months or weeks.

Why is identifying seasonality important when forecasting time series data?

Consider how ignoring regular cycles might confuse your forecast.