Time series Forecasting (TREND ,SEASONALITY)

Data Science, AI and ML
,
#1

Trend

The general direction of the data over time. For example, if we are looking at the height of a newborn baby, their height will follow an upward trend into their youth. On the other hand, someone on a successful weight loss program will see their weight follow a downward trend over time.

Seasonality + Cycles

Any seasonal or repeating patterns with a fixed frequency. Could be hourly, monthly, daily, annually, etc. One example of this is that Winter Jacket sales increase in the winter months and decrease in the summer months. Another example of this could be the balance of your bank account. In the 10 days at the start of every month, your balance follows a downward trend as you pay monthly rent, utilities, and other bill payments.

Irregularities + Noise

This is any large spikes or troughs in the data. One example of this could be your heart rate when you run the 400-meter dash. When you start the race your heart rate is similar to what it has been throughout the day, but during the race, it spikes to a much higher level for a small period of time before returning to a normal level.