Message:

<p>[QUOTE=&quot;calcol, post: 2807961, member: 77639&quot;]There are a number of ways of handling the ends of a data series with regard to averaging.&nbsp; Let&#039;s assume no data will be used before the start or after the end, and if the end is the present date, obviously no data will be available beyond it.&nbsp; If the smoothing interval is 6, then the first and last three data points are problematic.&nbsp; One way is to simply use them as is.&nbsp; Another is to use 1/2 of the interval, averaging data points away from the ends (i.e. toward the middle).&nbsp; So for a smoothing interval of 6, the first data point would be replaced by the average of it and the next two data points; the last would be replaced by the average of it and the two previous data points.&nbsp; Another approach is to use the full number (six in the example) going forward from the first data point and back from the last data point.&nbsp; Any way you do it, the beginning and end of a smoothed data series will be a little noisy or be overly biased by its neighbors going forward (for first few data points) or going backward (for last few data points).</p><p><br /></p><p>If the averaging interval is much smaller than the total span, which is usual and proper, then the smoothed curve should thread through the middle of a plot of the unsmoothed data.&nbsp; However, if there are a few extreme outliers in the raw data and they are extreme in the same direction, the smoothed curve may float above or below the majority of the raw data.&nbsp; There are special techniques for eliminating or decreasing the effect of&nbsp; outliers.</p><p><br /></p><p>Cal[/QUOTE]</p><p><br /></p>

Your name or email address:

Do you already have an account?

• No, create an account now.
• Yes, my password is:
• Forgot your password?

Stay logged in