Time series analysis patterns in Pandas - Step-by-Step Execution

Concept Flow - Time series analysis patterns

Load time series data

↓

Check data frequency

↓

Identify trend pattern

↓

Identify seasonality

↓

Detect noise/randomness

↓

Apply smoothing or decomposition

↓

Analyze residuals and patterns

↓

Make forecasts or insights

This flow shows how to analyze time series data step-by-step: loading data, checking frequency, spotting trend and seasonality, handling noise, and preparing for forecasting.

Execution Sample

Pandas

import pandas as pd
import matplotlib.pyplot as plt

data = pd.Series([10,12,15,14,13,16,18,20,19,21],
                 index=pd.date_range('2023-01-01', periods=10))
rolling_mean = data.rolling(window=3).mean()
rolling_mean.plot()
plt.show()

This code creates a simple time series, calculates a 3-day rolling average to smooth data, and plots the smoothed series.

Execution Table

Step	Action	Data/Variable	Result/Output
1	Create time series data	data	[10,12,15,14,13,16,18,20,19,21] with dates 2023-01-01 to 2023-01-10
2	Calculate rolling mean with window=3	rolling_mean	[NaN, NaN, 12.33, 13.67, 14.00, 14.33, 15.67, 18.00, 19.00, 20.00]
3	Plot rolling mean	plot	Line graph showing smoothed trend over dates
4	Exit	-	Rolling mean calculated and plotted

💡 Rolling mean calculation ends after processing all data points; first two are NaN due to window size.

Variable Tracker

Variable	Start	After Step 1	After Step 2	Final
data	empty	[10,12,15,14,13,16,18,20,19,21]	[10,12,15,14,13,16,18,20,19,21]	[10,12,15,14,13,16,18,20,19,21]
rolling_mean	empty	empty	[NaN, NaN, 12.33, 13.67, 14.00, 14.33, 15.67, 18.00, 19.00, 20.00]	[NaN, NaN, 12.33, 13.67, 14.00, 14.33, 15.67, 18.00, 19.00, 20.00]

Key Moments - 3 Insights

Why are the first two values of rolling_mean NaN?

Because the rolling window size is 3, the first two data points don't have enough previous values to calculate the average, so they are NaN as shown in execution_table step 2.

What does the rolling mean help us see in the time series?

It smooths out short-term fluctuations (noise) to reveal the underlying trend, as seen in the plotted line in execution_table step 3.

Why do we use a date range as the index for the series?

Using dates as the index helps pandas recognize the data as time series, enabling time-based operations like rolling windows and plotting with correct time labels.

Visual Quiz - 3 Questions

Look at the execution_table step 2, what is the rolling mean value at the 5th data point (index 4)?

A 14.00

B 13.67

C 12.33

D NaN

💡 Hint

Check the rolling_mean values in execution_table step 2, column 'Result/Output' at position 5.

At which step does the code produce the plot showing the smoothed trend?

A Step 1

B Step 3

C Step 2

D Step 4

💡 Hint

Refer to execution_table 'Action' column to find when plotting happens.

If we change the rolling window from 3 to 5, how will the rolling_mean values change?

A No NaN values, same smoothing

B Fewer NaN values, more noise

C More NaN values at the start, smoother trend

D Rolling mean will be the same

💡 Hint

Think about how window size affects the number of initial NaNs and smoothing effect.

Concept Snapshot

Time series analysis patterns:
- Use pandas Series with datetime index
- Identify trend by smoothing (rolling mean)
- Rolling window size controls smoothness and NaNs
- Plot to visualize patterns
- Helps separate trend, seasonality, noise

Full Transcript

This visual execution shows how to analyze time series data using pandas. We start by creating a time series with dates as index. Then, we calculate a rolling mean with a window of 3 to smooth the data, which helps reveal the trend by reducing noise. The first two rolling mean values are NaN because there are not enough points to average. Finally, we plot the smoothed series to see the trend visually. This process is a basic pattern in time series analysis to understand data behavior over time.