Raj Kumar Gupta PGP08042 Vaibhav Gangwar PGP08194 Varun Parvathaneni PGP08195 Time series model said to follow a random walk as the differences from one observation to the next observation are random In a random walk model, the time series itself is not random, however, the first differences of time series are random.
Xt−Xt−1=et
Advantages:
• Stock and commodity prices follow a random walk since
today’s stock price is equal to yesterday’s stock price plus a random shock • the past movement or trend of a stock price or market cannot be used to predict its future movement Time series model said to follow a random walk as the differences from one observation to the next observation are random In a random walk model, the time series itself is not random, however, the first differences of time series are random.
as moving average (MA) components • AR components capture the "change since last time“ while MA components capture smoothed trends in the data • Hence more flexible as compared to other statistical models Results
Models MAE RMSE MAPE
Neural Network 0.026518 0.033505 2.153416
ARIMA 0.027091171 0.034169801 2.077353269
Random Walk 0.028459 0.036626 3.635207
• From the above table, we can see that Neural Network has lowest error • Thus Neural Network model is best Fit for prediction of Gold prices Forecast Plots