In this video, we will be discussing about the smoothing techniques which are used in time series analysis. So, when time series data is collected over a long period of time, the data might suffer from certain variations or random variations to remove or to remove the effect of these random variations the techniques that are used are called smoothing techniques. So, when data is collected over time, it displays different variations or different random variation smoothing techniques can be used to reduce or cancel the effect of these variations, when properly applied these techniques smooths out the random variation in the time series data to reveal the underlying trends. So, our time series is a sequence of observations which are ordered in time inherent in the collection of data taken over time in some form of the Mediation there exists methods for reducing of cancelling the effect. Due to these variations.

These methods are called the smoothing techniques. The first method of that the first method that we'll be discussing about are the moving average methods moving average is a summary measure of the movements of the time series variables, which reduces the distortions by evening of the fluctuations of the time series example the stock market prices, this method is easy to calculate and fairly effective. In order to remove fluctuations from the time series analysis, we use moving average methods and exponential methods or moving average of period k is a series of arithmetic means each of K successive observations of the series we start with the first k observations and take their arithmetic mean. At next step we believe the first observation and include k plus one observation and take their arithmetic me this process is continued until we arrive at the last key observation each of those are admitted in means will correspond to the midpoint of the time interval covered in the calculation of the average.

So, when case on each of those arithmetic mean will correspond to corresponds to the given time series, but when case even the moving average values follow midway in between two successive tabulated time series to correspond to the moving average values, the given time periods we have to consider a subsequent to item moving average. So, there can be a simple moving average where the weights given to each of the k values are equal to one there can also be weighted moving average where we assign different weights to different time period values where the summation of weights are equal to one. So, here we have considered the time series data captured in the stock market prices. So, if we use a moving average methods moving average methods give equal weighted to all the prices in the selected time period and ignores the previous historic prices if prices taken as the time dependent variable Now, the next methods exponential methods that is exponential moving average exponential moving average gives more weightage to recent prices and less to older prices.

If prices considered as a time dependent variable, unlike moving average, the older share of the prices never goes away that's potential moving average. The equations for Exponential Moving Averages current exponential moving average is equal to current prices minus previous days into exponent plus previous jersey, their exponent is equal to two divided by number of days or months or years involved in exponential moving average and previous is equal to five days or month or year moving average now exponential moving average smooths out random calculations, but the season and trend competence still remain double exponential smoothing is better at handling trends. Triple exponential smoothing is better at handling parabolic trends, parabolic trends, double exponential smoothing can be used in data sets, which involve seasonality. Exponential Smoothing is one of the most popular techniques. You will Flexibility is in calculation and is good performance. Now let's move to these two smoothing techniques that is halts exponential smoothing and winters exponential smoothing.

Exponential Smoothing is used to remove random fluctuations and trend components. Winter exponential smoothing is used to remove all the components of the MCs so we will be learning to hear in this video Let's end this video over here. Thank you Goodbye See word for the next week. You