What is ARIMA in Python?

What is ARIMA?

The ARIMA modelstands for AutoRegressive Integrated Moving Average allows us to forecast a time series using the series’ past values.

A time series is a collection of data points collected at constant time intervals. Time series are used to forecast future values based on previous values.

A stationary time series is one whose statistical properties (mean, variance, autocorrelation, etc.) are all constant over time. A non-stationary series is one whose statistical properties change over time.

ARIMA model

An ARIMA model is characterized by three terms: p, d, q.

• p is the order of the ARauto-regressive part of the model term.

• q is the order of the MAmoving average term.

• d is the number of differencingthe number of past time points to subtract from the current value required to make the time series stationary.

ARIMA model in python

In this example, we will predict the next 10 days of stock prices from a given data of 100 days.

Step 1

Import the relevant libraries to perform time series forecasting:

import numpy as np, pandas as pd
import statsmodels.tsa.stattools as ts
from statsmodels.tsa.arima_model import ARIMA
import matplotlib.pyplot as plt

Step 2

Upload the relevant dataset using pandas.read_csv() method:

file = pd.read_csv("data.csv")
// prices is a field in .csv file containing all stock prices.
stock_price = df['prices']

You can view this data in stock_price using the plt.plot() method:

plt.plot(stock_price)

Below is the code to an output the variation in stock price for the last 100 days. It also contains a .csv file with sample stock prices.

Step 3

Initialize the ARIMA model and set the values of p, d, and q as 1, 1, and 2.

model = ARIMA(stock_price, order=(1,1,2))
model_fit = model.fit(disp=0)
// summary provides a detailed summary of the time series model
print(model_fit.summary()) 

Step 4

Let’s predict the next 10 values and plot them on a graph:

pred = model_fit.predict(100,109,typ='levels')
// 100-109, refers to the next 10 values after the value at 99th index. 
newarr = []
for i in price:
  newarr.append(i)


for x in pred:
  newarr.append(x)

plt.plot(newarr)

Complete code