i used z score to normalize my actual data to predict using NNtoolbox. can any one tell me how to denormalize my predicted data.

this is my actual data. 0.000328137 0.000261894 0.000323088 0.000241772 0.000336459 0.000219204 0.000280989 0.000329144 0.000261423 0.000219585 0.000214844 0.000213005 0.000234333 0.000276782 0.000221198 0.000204171 0.000222497 0.00021443 0.000168888 0.000226346 0.000170379 0.000241959 0.000247086 0.000378657 0.000298095 0.000322626 0.000187401 0.000256619 0.000226761 0.000183668 0.000283274 0.00018668 0.000282647 0.000247847 0.000271565 0.000210266 0.000238253 0.000144414 0.000142482 0.000174384 0.000122349 0.000163343 0.000269351 0.00025746 0.000279306 0.000175955 0.000275318 0.000243587 0.0002472 0.000186734 0.000201168 0.000198873 0.000259911 0.000230289 0.000216636 0.00018573 0.000251924 0.000295206 0.000351549 0.000257067 0.000269793 0.00017142 0.000245543 0.000238745 0.000240741 0.000235238 0.000294257 0.000303185 0.000277074 0.000243998 0.000229013 0.000262289 0.000254646 0.000212603 0.000339615 0.000211053 0.000226884 0.000352547 0.000255893 0.000242007 0.000236975 0.000248978 0.000205545 0.000269982 on normalizing using zscore i get 1.677761461 0.369003586 1.578008755 -0.028555105 1.842178578 -0.474419473 0.746262038 1.697646805 0.359698075 -0.466892085 -0.560569537 -0.59690252 -0.175516979 0.663134783 -0.435034046 -0.771425302 -0.409369803 -0.568748904 -1.46850883 -0.333325405 -1.439051257 -0.024850682 0.076443065 2.675881231 1.084223969 1.568881057 -1.102748907 0.264775937 -0.325116403 -1.176511381 0.791396729 -1.117003528 0.779009138 0.091468205 0.560062914 -0.651016733 -0.098079718 -1.952039773 -1.990210149 -1.359924779 -2.387976287 -1.578070848 0.516321085 0.281401388 0.713011136 -1.328896531 0.634210647 0.007303711 0.078695354 -1.115936654 -0.830755343 -0.876097482 0.329825607 -0.255413977 -0.525155252 -1.135762726 0.172017181 1.027146218 2.140310552 0.273636917 0.525063524 -1.4184843 0.045948253 -0.088359312 -0.048914614 -0.157646842 1.008387019 1.184776831 0.668913683 0.015423807 -0.28063369 0.376797692 0.225805405 -0.604844803 1.904531428 -0.635458156 -0.322686302 2.160018083 0.25043241 -0.023902349 -0.123329066 0.113813285 -0.74427929 0.528797582

now from nftool i get my predicted data to be

1.677761461 0.369003586 1.578008755 -0.028555105 1.842178578 -0.474419473 0.746262038 1.697646805 0.359698075 -0.466892085 -0.560569537 -0.59690252 -0.175516979 0.663134783 -0.435034046 -0.771425302 -0.409369803 -0.568748904 -1.46850883 -0.333325405 -1.439051257 -0.024850682 0.076443065 2.675881231 1.084223969 1.568881057 -1.102748907 0.264775937 -0.325116403 -1.176511381 0.791396729 -1.117003528 0.779009138 0.091468205 0.560062914 -0.651016733 -0.098079718 -1.952039773 -1.990210149 -1.359924779 -2.387976287 -1.578070848 0.516321085 0.281401388 0.713011136 -1.328896531 0.634210647 0.007303711 0.078695354 -1.115936654 -0.830755343 -0.876097482 0.329825607 -0.255413977 -0.525155252 -1.135762726 0.172017181 1.027146218 2.140310552 0.273636917 0.525063524 -1.4184843 0.045948253 -0.088359312 -0.048914614 -0.157646842 1.008387019 1.184776831 0.668913683 0.015423807 -0.28063369 0.376797692 0.225805405 -0.604844803 1.904531428 -0.635458156 -0.322686302 2.160018083 0.25043241 -0.023902349 -0.123329066 0.113813285 -0.74427929 0.528797582

i owuld like to know how to denormalize this data.

[Greg Heath]

Your predicted data is EXACTLY the same as your normalized data!

Answer 1

[Greg Heath]

If you normalize your target data using

[ t meanT stdT] = zscore(T)

and the normalized output data is

y = sim(net,x); or y = net(x);

then

Y = meanT + stdT.*y;

Hope this helps.

Greg

P.S. If T is multidimensional use repmat so that meanT and stdT can be combined with y

Answer 2

[Star Strider]

If you used mapstd, use the 'reverse' option.

Note that (from the documentation): ‘ For most networks, including feedforwardnet, these steps are done automatically, so that you only need to use the sim command. ’