股市大盤指數漲跌預估

Scheme:weka.classifiers.bayes.NaiveBayes Relation: TWN_STOCK_IDX-

weka.filters.unsupervised.attribute.Remove-R1 Instances: 64 Attributes: 5 開盤指數差 最高指數差 最低指數差 收盤指數差 指數漲幅差 Test mode:10-fold cross-validation === Classifier model (full training set) === Naive Bayes Classifier Class Attribute N Y (0.52) (0.48) =============================== 開盤指數差 mean 23.2525 -9.3239 std. dev. 40.2533 44.1367 weight sum 33 31 precision 3.441 3.441 最高指數差 mean 0 14.108 std. dev. 35.0766 49.3931 weight sum 33 31 precision 4.0124 4.0124 最低指數差 mean -0.6221 14.2389 std. dev. 46.0191 30.8398 weight sum 33 31 precision 3.4217 3.4217 收盤指數差 mean -22.739 37.6398 std. dev. 41.3947 37.9565 weight sum 33 31 precision 3.9287 3.9287 === Stratified cross-validation === === Summary === Correctly Classified Instances 54 84.375 % Incorrectly Classified Instances 10 15.625 % === Confusion Matrix === a b <-- classified as 27 6 | a = N 4 27 | b = Y

=== Re-evaluation on test set === User supplied test set Relation: TWN_STOCK_IDX2-

weka.filters.unsupervised.attribute.Remove-R1

-weka.filters.unsupervised.attribute.Remove-R5 Instances: unknown (yet). Reading incrementally Attributes: 5 === Predictions on test set === inst#, actual, predicted, error, probability distribution 1 1:N 1:N *0.979 0.021 2 2:Y 1:N + *0.958 0.042 3 2:Y 2:Y 0.256 *0.744 4 2:Y 2:Y 0.203 *0.797 5 1:N 1:N *0.969 0.031 6 2:Y 2:Y 0.418 *0.582 7 2:Y 2:Y 0.199 *0.801 8 1:N 1:N *0.738 0.262 9 2:Y 2:Y 0.097 *0.903 10 1:N 1:N *0.604 0.396 11 1:N 1:N *0.759 0.241 12 2:Y 2:Y 0.18 *0.82 13 1:N 1:N *0.857 0.143 14 2:Y 2:Y 0.21 *0.79 15 1:N 1:N *0.909 0.091 16 2:Y 2:Y 0.282 *0.718 === Summary === Correctly Classified Instances 15 93.75 % Incorrectly Classified Instances 1 6.25 % Kappa statistic 0.875 Mean absolute error 0.2492 Root mean squared error 0.3276 Total Number of Instances 16 === Confusion Matrix === a b <-- classified as 7 0 | a = N 1 8 | b = Y