Why moving average plot is not starting as the original signal

6 次查看(过去 30 天)
Hi, I am trying to analyze a stock data with only two columns i.e dates and value of stocks Both of these columns have length=63.I am trying to calculate the moving average to smooth the curve.I use the following code and get the out put as shown below
window=5;
mask=ones(1,window)/window;
quarter=conv(stocks,mask,'same');
plot(dates,stocks)
hold on
plot(dates,quarter)
ylim([7300,7700])
legend('input data' ,'filtered data')
Why the filtered data curve is not starting from the point where the original input data is starting? How to fix the plot? Is my method to get the moving average right? Is there a better way? thanks

回答(1 个)

Walter Roberson
Walter Roberson 2018-4-4
You are getting effects from the zero padding. See the 'valid' option instead of 'same'
Or consider using
quarter = movavg(stocks(:), 'simple', window) .';
  2 个评论
rohail khan
rohail khan 2018-4-4
hi, Thank you so much for quick reply. I tried the code but now its giving me the following error: 'Undefined function 'movavg' for input arguments of type 'double'
Please can you explain how to use the 'movavg' function.
Thanks
Walter Roberson
Walter Roberson 2018-4-4
I forgot that movavg is from the Financial Toolbox. You might want movmean() from R2016a or later.

请先登录,再进行评论。

类别

Help CenterFile Exchange 中查找有关 Data Preprocessing 的更多信息

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by