Air-Passengers-Time-Series-Analysis

Forecasting number of passengers for airlines using ARIMA model in python. You can view the project demo on YouTube.

Table of contents

• General info
• Screenshots
• Demo
• Methods and Tools
• Code Examples
• Status
• Contact
• Reference

General info

Project was Time Series Analysis, used ARIMA method to build the model. Major steps involved were as follow :

• STEP: 1 - Data Cleaning and Analysis
• STEP: 2 - Checking Stationarity (ADF Test)
• STEP: 3 - Transformation
• STEP: 4 - Differencing
• STEP: 5 - Time Series Components
• STEP: 6 - Finding ACF and PACF
• STEP: 7 - ARIMA Modeling
• STEP: 8 - Forecast

Demo

The entire demo of the project can be found on YouTube.

Screenshots

Methods and Tools

• Python
• Advanced Excel
• ARIMA
• Augmented Dickey-Fuller Test
• ACF and PACF
• Statsmodels

Code Examples

# Code for Augemneted Dickey-Fuller Test and Rolling Mean to check stationarity  

def stationarity(timeseries):
    
    rolmean=timeseries.rolling(window=12).mean()
    rolstd=timeseries.rolling(window=12).std()
    
    plt.figure(figsize=(20,10))
    actual=plt.plot(timeseries, color='red', label='Actual')
    mean_6=plt.plot(rolmean, color='green', label='Rolling Mean') 
    std_6=plt.plot(rolstd, color='black', label='Rolling Std')
    plt.legend(loc='best')
    plt.title('Rolling Mean & Standard Deviation')
    plt.show(block=False)
    
    print('Dickey-Fuller Test: ')
    dftest=adfuller(timeseries['Passengers'], autolag='AIC')
    dfoutput=pd.Series(dftest[0:4], index=['Test Statistic','p-value','Lags Used','No. of Obs'])
    for key,value in dftest[4].items():
        dfoutput['Critical Value (%s)'%key] = value
    print(dfoutput)

Status

Project is: finished.

Contact

If you loved what you read here and feel like we can collaborate to produce some exciting stuff, or if you just want to shoot a question, please feel free to connect with me on email or LinkedIn

Reference

• Edureka