<html><head></head><body><div style="color:#000; background-color:#fff; font-family:HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif;font-size:16px"><div id="yiv2870226355yui_3_16_0_1_1450708156377_6602" class="">Dear all</div><div id="yiv2870226355yui_3_16_0_1_1450708156377_6638" class=""><br class="" id="yui_3_16_0_1_1450764885817_3670"></div>
<div id="yiv2870226355yui_3_16_0_1_1450708156377_6600" class="">I am using spatio-temporal <span class="" id="yiv2870226355yui_3_16_0_1_1450708156377_6605" style=""> </span>regression kriging for estimation of my air temperature
data. After removing the trend in data with a linear model, I calculate the
residuals. I found empirical semivariogram and fit a sum metric model to it ( I
attach the plot of them). It shows spatio-temporal dependency among the
residuals. But, when I used the variogram for spatio-temporal kriging estimation
of the residuals and add them back to the linear model result, it increases the
error of the estimation. </div>
<div id="yiv2870226355yui_3_16_0_1_1450708156377_6607" class="">Shouldn’t this method decrease the error of estimation?</div>
<div id="yiv2870226355yui_3_16_0_1_1450708156377_6609" class="">Does anybody know why this is happened?</div>
<div id="yiv2870226355yui_3_16_0_1_1450708156377_6611" class="">What should I do to resolve this problem?</div><div id="yiv2870226355yui_3_16_0_1_1450708156377_6634" class=""><br class="" id="yui_3_16_0_1_1450764885817_3678"></div>
<div id="yiv2870226355yui_3_16_0_1_1450708156377_6613" class="">I really appreciate any help</div>
<div id="yiv2870226355yui_3_16_0_1_1450708156377_6615" class="">Thanks in advance</div><div id="yiv2870226355yui_3_16_0_1_1450708156377_6636" class=""><br class="" id="yui_3_16_0_1_1450764885817_3683"></div>
<div id="yiv2870226355yui_3_16_0_1_1450708156377_6617" class=""><div id="yui_3_16_0_1_1450764885817_3729" dir="ltr">Shohreh</div></div></div></body></html>