polyplot

This function simply plots a polynomial fit to scattered x,y data. This function can be used to easily add a linear trend line or other polynomial fit to a data plot.

Syntax
polyplot(x,y)
polyplot(x,y,n)
polyplot(...,'LineProperty',LineValue,...)
polyplot(...,'error')
polyplot(...,'error','ErrorLineProperty',ErrorLineValue)
[h,p,delta] = polyplot(...)

Description
polyplot(x,y) places a least-squares linear trend line through scattered x,y data.

polyplot(x,y,n) specifies the degree n of the polynomial fit to the x,y data. Default n is 1.

polyplot(...,'LineProperty',LineValue,...) formats linestyle using LineSpec property name-value pairs (e.g., 'linewidth',3).

polyplot(...,'error') includes lines corresponding to approximately +/- 1 standard deviation of errors delta. At least 50% of data should lie within the bounds of error lines.

polyplot(...,'error','ErrorLineProperty',ErrorLineValue) formats error lines with line property name-value pairs. All arguments occurring after 'error' are assumed to be error line property specifications.

[h,p,delta] = polyplot(...) returns handle(s) h of plotted line(s), coefficients p of polynomial fit p(x), and error estimate delta, which is the standard deviation of error in predicting a future observation at x by p(x). Assuming independent, normal, constant-variance errors, y +/- delta contains at least 50% of the predictions of future observations at x.