Aprenda Visualizing Seasonality with Heatmaps

Deslize para mostrar o menu

Seasonality is a fundamental concept in time series analysis, describing patterns that repeat at regular intervals, such as daily, monthly, or yearly cycles. Recognizing these patterns is crucial for accurate forecasting and deeper understanding of the underlying processes in your data. Visualizing seasonality helps you spot recurring behaviors, outliers, and shifts in temporal dynamics that may not be obvious through raw numbers or simple line plots. One powerful way to visualize and interpret seasonality is with a heatmap, which can highlight cyclical trends and anomalies across different time scales.


              123456789101112131415161718192021222324252627
            
import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt

# Generate a sample time series dataset with clear seasonality
np.random.seed(42)
dates = pd.date_range(start="2018-01-01", end="2022-12-31", freq="D")
seasonal_pattern = 10 + 5 * np.sin(2 * np.pi * dates.dayofyear / 365)
noise = np.random.normal(0, 2, len(dates))
values = seasonal_pattern + noise

df = pd.DataFrame({"date": dates, "value": values})
df["year"] = df["date"].dt.year
df["month"] = df["date"].dt.month

# Group by year and month, then calculate monthly averages
monthly_avg = df.groupby(["year", "month"])["value"].mean().unstack()

# Create a heatmap to visualize seasonality
plt.figure(figsize=(10, 6))
sns.heatmap(monthly_avg, cmap="YlGnBu", annot=True, fmt=".1f", cbar_kws={'label': 'Average Value'})
plt.title("Seasonal Heatmap: Average Value by Month and Year")
plt.xlabel("Month")
plt.ylabel("Year")
plt.tight_layout()
plt.show()

The heatmap you generated displays years on the y-axis and months on the x-axis, with each cell showing the average value for a specific month and year. By examining the color intensity and annotated values, you can quickly identify recurring seasonal patterns. For instance:

Warmer colors consistently appearing in the summer months across all years signal a regular seasonal increase;
Cooler colors in winter months suggest a recurring dip;
Outliers or anomalies may show up as unexpected color shifts that break the typical pattern.

This visual approach makes it much easier to interpret cyclical behaviors, compare seasonality across years, and spot any changes or disruptions in the pattern over time.

Tudo estava claro?

Obrigado pelo seu feedback!

Seção 1. Capítulo 6

Pergunte à IA

Pergunte o que quiser ou experimente uma das perguntas sugeridas para iniciar nosso bate-papo

Seção 1. Capítulo 6