- Financial Time Series Data
- Exploring Time Series Data in R
- Plotting Time Series in R
- Handling Missing Values in Time Series
- Creating a Time Series Object in R
- Check if an object is a time series object in R
- Plotting Financial Time Series Data (Multiple Columns) in R
- Characteristics of Time Series
- Stationary Process in Time Series
- Transforming a Series to Stationary
- Time Series Transformation in R
- Differencing and Log Transformation
- Autocorrelation in R
- Time Series Models
- ARIMA Modeling
- Simulate White Noise (WN) in R
- Simulate Random Walk (RW) in R
- AutoRegressive (AR) Model in R
- Estimating AutoRegressive (AR) Model in R
- Forecasting with AutoRegressive (AR) Model in R
- Moving Average (MA) Model in R
- Estimating Moving Average (MA) Model in R
- ARIMA Modelling in R
- ARIMA Modelling - Identify Model for a Time Series
- Forecasting with ARIMA Modeling in R - Case Study
- Automatic Identification of Model Using auto.arima() Function in R
- Financial Time Series in R - Course Conclusion

# Estimating AutoRegressive (AR) Model in R

We will now see how we can fit an AR model to a given time series using the `arima()`

function in R. Recall that AR model is an `ARIMA(1, 0, 0)`

model.

We can use the `arima()`

function in R to fit the AR model by specifying the `order = c(1, 0, 0)`

.

We will perform the estimation using the `msft_ts`

time series that we created earlier in the first lesson. If you don't have the `msft_ts`

loaded in your R session, please follow the steps to create it as specified in the first lesson.

Let's start by creating a plot of the original data using the `plot.ts()`

function.

```
> plot.ts(msft_ts, main="MSFT prices", ylab="Prices")
```

As of now we are not worried about whether an AR model is best suited for this data or not. Our objective is to understand the process of fitting the AR model to this data.

We will fit the AR model to this data using the following command:

```
msft_ar <- arima(msft_ts , order = c(1, 0, 0))
```

The output contains many things including the estimated slope (ar1), mean (intercept), and innovation variance (sigma^2) as shown below:

```
> msft_ar
Call:
arima(x = msft_ts, order = c(1, 0, 0))
Coefficients:
ar1 intercept
0.9815 56.2260
s.e. 0.0113 2.2359
sigma^2 estimated as 0.5891: log likelihood = -292.55, aic = 591.1
>
```

The `msft_ar`

object also contains the residuals (εt ). Using the `summary()`

function, you can see that the object contains a time series of residuals.

> summary(msft_ar) Length Class Mode coef 2 -none- numeric sigma2 1 -none- numeric var.coef 4 -none- numeric mask 2 -none- logical loglik 1 -none- numeric aic 1 -none- numeric arma 7 -none- numeric residuals 252 ts numeric call 3 -none- call series 1 -none- character code 1 -none- numeric n.cond 1 -none- numeric nobs 1 -none- numeric model 10 -none- list

You can extract the residuals using the `residuals()`

function in R.

residuals <- residuals(msft_ar)

Once you find the residuals εt, the fitted values are just X̂t\=Xt−εt. In R, we can do it as follows:

msft_fitted <- msft_ts - residuals

We can now plot both the original and the fitted time series to see how close the fit is.

ts.plot(msft_ts) points(msft_fitted, type = "l", col = 2, lty = 2)

### Calculating Residuals Manually from the Fitted Model (Advanced Content)

The AR(1) model is expressed as follows:

**Xt = Ar * Xt-1 + εt**

So,

**εt = Xt - Ar * Xt-1**

Where Ar is the estimated autoregressive part in the fitted model.

n=252 e=rep(1,n) e[1]=0 for (t in (2 : n)){ e[t] = msft_ts [t]-coef(msft_ar )[1]*msft_ts [t-1] }

Suppose the model was ARIMA(1,0,1)

**εt = Xt - Ar * Xt-1 - Ma * Xt-1**

> n=252

e=rep(1,n) e[1]=0 for (t in (2 : n)){ e[t] = msft_ts [t]-coef(msft_ar )[1]*msft_ts [t-1]-coef(msft_ar )[2]*e[t-1] }