Is it possible to use the neural network toolbox to create scenarios for the output? As I understand, the sim function gives a forecast of the output. However, what I want is to create several scenario paths to capture the risk in the future values of the output variable.
Thanks for help!

1 件のコメント

Greg Heath
Greg Heath 2014 年 6 月 9 日
Insufficient information and explanation.

サインインしてコメントする。

 採用された回答

Greg Heath
Greg Heath 2015 年 2 月 1 日

0 投票

Add random noise to the test data to get performance vs snr.
Better yet, systematically add noise to all the data to get a number of nets, each of which was designed at a constant SNR level. Then, for each net, test with each of the noise levels to get a performance matrix of size SNRtrain x SNRtest.

その他の回答 (0 件)

カテゴリ

ヘルプ センター および File ExchangeDeep Learning Toolbox についてさらに検索

質問済み:

2011 年 3 月 17 日

回答済み:

2015 年 2 月 1 日

Community Treasure Hunt

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

Start Hunting!

Translated by