iVBORw0KGgoAAAANSUhEUgAAAXoAAAA0CAIAAAAL52RvAAAABGdBTUEAALGPC/xhBQAAAAlwSFlzAAAScQAAEnEB89x6jgAAABJ0RVh0U29mdHdhcmUAR3JlZW5zaG90XlUIBQAADRxJREFUeF7tnd+OFccRh/0GzhvEb4DfwH4D+w3sN0jewFxxhxIpcGEJCS4s+WKJsYR9w14kYq8WIWE2CgqKMVqL4FUQ8kIisCVL5NtTvZXa6j/TZ2fm7J51fWqNTtdU93QPU7+pnjNneetNEATBSgi5CYJgRYTcBEGwIk5Gbv7y5D83Hu2nSrAOPHny5O7du7u7u6keBMuzarl5/OLn97/49q0/fUNJpmAdCLkJxrNSubn37NVvPv2baM2ycvPjT7+c395DqhCsZApWSMhNMJ6Vys0n2z+o1iwlN1anWIgla7BCQm6C8ayH3Lz96Y62Crk5EUJugvGsh9zYViE3J0LITTCekJugi5CbYDwhN0EXITfBeEJugi5CboLxhNwEXYTcBOMZkJsff/rlj988+3jz+/e/+Na+MkN59/OH2Nl779mr5D3EycoN4zy/vff72/9iLmz5/OWjF2lfMETITTCeAbn54KvvbKjXyjvXHhDAgy/gLSU3f7j3b3SBgq7ZVmgcSmHLoABdffDcaaUW7PSAqibXEuz965P/Skmmxe8wRLlEvJJ1AbvUv92zok2sdvNZ+0mmCnaEM70GGXITjGdAbj7a3HXx2SiEbmpWYSm5sZ7t8t6f/5naZBCHTq2KBZ+GLljNRQTxRGLUIiW5LrB2TmCy1kEN1f+31/6erEffNvrd7SfJmuHm2HPEYxByE4xnQG6463LRE8+EHGJhCxYbD1LaiQatrHOyVrCe7dKQm3Of/yN3luLsRGxqk2GdmXWeKLkBWHkalGCwg+QUJesiv1M7pXZuUSLrttbZzf7+PocAPiRTk9evX4s/JFOdTjdFBsMhUn3Bsp3s7e25TsTCNtVLFH0GW4GM2Z09LNYo1R7Ef0LGPip2CjKh3ORKUSs2RC3ucFRtCsNnF6i1fnJtcsUlFDce7du97d++ow7W2YmFPQmsWPMUjBOuDpTaFMYzq9wQCTdv3rx8+fIFA9XNzU0X7QLGra2tK1euJNdDsOzs7CSnDPG5detWqjd5+PCh+Lspi5HRpnqT27dvi7/t5LPPPsPCNtVLFH0WPV24f/9+qpeQI7JN9QXSUI1S7UFOwsWLFxsyxy5xvnPnTjLVGSs37nKfUG6UpQ6huDAmU0g7jmIVh3hO1qPkckNah8QwknvPXrHNVYA1kTq3Vzc2hXFZEtC/7qW4JRXHZcy6F21KO2ZgPrkhfuR6LcK1zkWfXBdwfWNMuxcQltaC6BTDI+2+cGEwR4BLly6Jc1FuYPBUoIk6qgnlph3808oNzoOjHXSwnFm5sTpiH4g4iFh1oxQ7d3JDVNsHukVc3pTrkWKF6eqD58lqcCfNjtAdZXBUY5hJbrglLi7sA4gHG0gciyQC+8bGRjId1Rr2Wn9SJHqQvcWYlFYwGBtWAWtyM9iJRL4wodwAelpM+kAOqsoiSCtnVBibOBT/cTmrsreYVemJaiig5czKjX2uVEttBKsmjDBZDdaBbnui2uVWRR0Bm7zQc02Viksqd1qKI5+QOeSGa1Qu1sYdG7sNLV1AuZRHwV8UB89kOgSjSlVjIpqVSILjPBetUz+NdU2jk5FyI93WVnOTyw1In6xtncbpHDvXp3A25ebYo/rgq++S1WDlJl/v1LAaUewWbHrSWHO56ciSyvZPitRIoCZhDrmRoILOe6PeS2uRI2j8ODkQI7HBluBJ1gyJLtxkeG7KB10sBsC20YnkZWzzTopS4ij6YAE9CUWxk4G58yP+tZOmp6vxjyui6XqQY6E4ToYanE25cd/pJGsFO6qimhxPbuzX25SiHNgUrD0vt2768OvHttpzTkYyudxoll67UedIatNzfTfCVe/JxXCVvXII6aQoN9pJMYZlajjwIe+kODZH0QcL8EG0DHKZFglwoxLn4lChR27UhxmJRf/5amlmkbMpN2QKtkmyVphJbtAXbUXJ11P2C6zG0yWB3uxTHlsk2ZmbyeVG79Kd1ysRLv498qSPhFJ9gVoat2WJZInMXCngoItFJzL+Y3RSlBJH0QcL8IEjivLmCxyZmlOWRbtRcgMbGxv4XL9+Xao9s8jpkhuu9S8fvTi/vcdNVd6j1eJeojslcmMFguLG7Ir9cmdCuQGrehwoWQ+xe3skw50HKStYRgmTy40EBnSupDQqamFjKYaQWORzcXUgd2x2SQxLRLkpH3RxtBP35EJTm1onPYFa9MEC8lkfUWn8C3JW3bwW7cbKDfOSI+KmTTTZ6WRAbh6/+Pnjze/dJd4op1Nu+su0cuNewLHv1Ljcx71uU8MtqSgr+w8t5pObVB9CL/Fa2FjU2Q5YLPJZcxMbMHID10VWrhRw0MVhJ3oU24m00pdQ8k6KUuIo+mCBVKk8yZKz6k5R7mYpnqsikjMyKhlezz+EoyU35P+1nxrVyhxy4x6CzCo3xee1x5YbsCsg+wWZndS57vdlnPL2vLI8FfPJTecdUqOi5yovhpBYUuUwpHVpJk1IWKQK4uCmfNBF1ommGD2diIVtqpco+mCBVFkgj71B+5ez6k6R+NTOm4wZ3EyL6DeDmgMuRVVuuN86rSEq0AsKd1Suey3Wh2pqX+J4cuNatQ8hWIFg2DrUwZLaH2WM3NjB299J2D7zxzo1mIu2kkL/ad/MzCc3nX2yfBD/HrmR12HBaplYUsWEmazmarrghrdo8f9OdFTiJk00P4Jat2xTvUTRBwukyiHu8fnccqMndqknxEpVbuwPf9ovm6gbpRaxwonIzfgUYIzcuBdwZNHkjJ0PX9x50DLr233K5HKjF26PfAjyrCR/oSZHbvs2y4DF0Y6Eq6yeiGpZmLjwlph3Uz7o4mgn8mAYZ4nbwU70oKleQlo5HyyQKoegp/JIRU7L3HKzlHNOVW7sBd1+Tc56nkK5oSTrcRkjN2Cby5m039M3Xrex2BcCXWn8uHRCJpcbkDhpvL3ikMAevNa5z0vPulASpG2qLCBWxSj+7qF1rhSwcPed6ETYOv+8E1EEmqR6CRHWomqkikGFG5FdS7khpO0F3Y7wfs+VyY1rMvK7m5FyYx/TyM+y7Jqo81mvbUKy6V4sYr7JbzbmkBuJDahFgkPVgTt548GBqpJdSYEYU+UQffzhtAk65QZ0Ii4fgbwTXX/ZNZdFQ9rJnxhT5Sg6C1lbrbfcJGsF67mU3BxvEdEjN278/Q9HioyUG6ZpX+ez6jP4uo3gxEVSJDsqytxLqjnkRtMQqP2emECyQqCBTVw5NQE6vH79ujjkHYo9VQ6RMUDeW7/cLNuJGPHPzycSg529uXIdHLgiNyB9CustN+0I7/c8hnCAC7aedMB9zZy/87IUI+UG3GuHWnpet3HP7HUA2K2Kzb2kmkNuQKMLWIygERxC4LMsT1wYa/ICiMvW1harCbbYtas8VQHZlSoGxAJSxdAvN7BUJ0xHh8oUdnZ2FjPe3dzcFCN7XWoDsitVMji69rlmcuOeFDSyA/fUs60g9sZO6Xxs4bSvs5WL8DEJzni5cVPQIk+O29ijoy+2STHrmYmZ5AaIK1kCFCFc88BjGSJPN3Kw1xYp4pAqHSwlNzWKnUBj1kwhnzLI3lQpoVqwZnID9s75zrUHeWAgSfkbgNxj5e8Hk1DkEZ4/78RTeiYfkReX80WBUzSK/l1ktv2tcCsu3xqHFsbLDeQ/QTjX8bqNE5T8GY0dG0mQnJY5mE9uBDIUUhKCk1s08cYHllHtw9GEENrY2MCZLZ+xpH0lcINU6YAB4O8if5JOFJSRkYvuMGs+17QSeg7NGcg7kYa1nhmbONQGaVnKOacqN/kLrCiI/AcGH3792L74XyvFb6BrP/zRMvib7GIpLiXc2k0KnjIR+UGGnUhtPTKJ3OSDGUy40A67jCrKk1PVkcvGBnPLTfBroCo33PMHpYGCT0ML8nyhtqzQUhSpPC3KS3I9Su2hSa0U059J5CbPtorHsrgTW0u+nJDNtKQKuQnGU5Ub4PrOX2O1hQyImCGQ7MpLi3vQoHBXL/pLqf1pmBuP9hutGguTPE1rlOKAJ5EbsP0MPoFyz7nQlLSjhP1nmmlJFXITjKclNwLXPRKgmQ4xT9hwC7U3Zz4TP+rD1U+QNy56/OmBbulKOqQQUbUbuFBsxYEGv6tiJAyvoVaMvDFgBiZt2Q6ugBowcj3i4JiZqUxQ5pisFThv6kxpn8bjEXITjGdYbs4SxCFxjnxIIaRZ3M2RC+RwFNUaZCtZ14eQm2A8vy65OUHsmm4wWzmFhNwE4wm5WQUsA+13TKvJp6Yl5CYYT8jNKmDhplrz3oiHzSdIyE0wnpCb2bn37JVNbdovXp9aQm6C8YTczAhrqPPbe1Zr1jS1gZCbYDwhNxODoCAr7x79i/FS3q7/x3Wnn5CbYDwhNxPjJEYLWjPH6zArI+QmGE/IzcQ4lZHy0ebu+uY1QshNMJ6Qm4nR957ZfrL9w9UHz9fxa++cly9fPn36NP/zUUHQT8hNEAQrIuQmCIIVEXITBMFKePPmf4GyDTAL6GK4AAAAAElFTkSuQmCC
Weekly Challenge #157
You are provided a dataset (Q2_variables.yxdb) that contains multiple variables. Select the ten (10) numeric variables with the highest Mean Decrease Gini coefficient from the variable importance plot. Use these variables to build a model to predict the target variable, [H0]. Compare two models: one based on all of the selected variables, and another that includes the selected variables except [F_38]. What is the effect of removing this variable [F_38] from the model? Provide the Chi-Sq effect as your answer.
Challenge 157 - Q2_variables.yxdb
Challenge 157 - Q2_variables.yxdb
Step 3 - Check for Correlations
C:\Users\Philip\AppData\Local\Temp\Engine_15412_46799beda7c646e097a18a06a953e544_\Engine_16788_5033779cafce4d4bbe1fdf3925f7bc9d_.yxdb
Single
Profile
Formula
[Name] in('H0', 'S_11', 'F_15', 'S_12', 'F_29', 'F_19', 'S_3', 'S_6', 'F_23', 'F_38', 'F_21')
Warn
Value - Descending
[FieldName] > [Name]
Custom
>
FieldName
True
fixed
2019-04-30 07:55:02
0
Name
2019-04-30 07:55:02
2019-04-30 07:55:02
[FieldName] > [Name]
Step 2 - Does removing F_38 make a difference?
Home
advanced
True
False
Logistic_Regression_All
H0
S_2,S_3,S_4,S_5,S_6,S_7,S_8,S_10,S_11,S_12,F_15,F_16,F_17,F_18,F_19,F_20,F_21,F_22,F_23,F_24,F_26,F_28,F_29,F_30,F_31,F_33,F_34,F_35,F_36,F_38,F_39,F_42,F_44,F_46,F_47,F_48,F_49,F_50,F_51
False
False
0.5
False
False
5
lambda_1se
False
1
0.5
logit
False
5
3
False
False
1
1x
Logistic_Regression_All
Home
advanced
True
False
Logistic_Regression_Subset_with_F38
H0
S_3,S_6,S_11,S_12,F_15,F_19,F_21,F_23,F_29,F_38
False
False
0.5
False
False
5
lambda_1se
False
1
0.5
logit
False
5
3
False
False
1
1x
Logistic_Regression_Subset_with_F38
C:\Users\Philip\AppData\Local\Temp\Engine_15412_46799beda7c646e097a18a06a953e544_\Engine_16788_5d9534176cc04d0a95ecf18d87757101_.yxdb
Single
Report
C:\Users\Philip\AppData\Local\Temp\Engine_15412_46799beda7c646e097a18a06a953e544_\Engine_16788_5a29f6b4aed542919feda1e55f17ef47_.yxdb
Single
Report
Diff between all predictors and top 10
C:\Users\Philip\AppData\Local\Temp\Engine_15412_46799beda7c646e097a18a06a953e544_\Engine_16788_1823a6b8c15b4660b845d8772fb8910f_.yxdb
Single
Report
Home
advanced
True
False
Logistic_Regression_Subset_with_F38
H0
S_3,S_6,S_11,S_12,F_15,F_19,F_21,F_23,F_29
False
False
0.5
False
False
5
lambda_1se
False
1
0.5
logit
False
5
3
False
False
1
1x
Logistic_Regression_Subset_with_F38
C:\Users\Philip\AppData\Local\Temp\Engine_15412_46799beda7c646e097a18a06a953e544_\Engine_16788_d6d9926848684106a72690d413da33cf_.yxdb
Single
Report
Diff between top 10 and top 10 less F_38
C:\Users\Philip\AppData\Local\Temp\Engine_15412_46799beda7c646e097a18a06a953e544_\Engine_16788_4d6ae328472e429db5dd68233bd5852f_.yxdb
Single
Report
The Effect of Removing the Variable F_38 from Log Reg Subset with F38 is Chi_Sq = 9.97 (p < 0.01) therefore significant
Step 1 - Find the most significant variables
Forest
H0
S_2 + S_3 + S_4 + S_5 + S_6 + S_7 + S_8 + S_10 + S_11 + S_12 + F_15 + F_16 + F_17 + F_18 + F_19 + F_20 + F_21 + F_22 + F_23 + F_24 + F_26 + F_28 + F_29 + F_30 + F_31 + F_33 + F_34 + F_35 + F_36 + F_38 + F_39 + F_42 + F_44 + F_46 + F_47 + F_48 + F_49 + F_50 + F_51
500
False
3
False
100
10
5
True
100
True
5.50
5.50
False
13.00
14.95
1x
10
C:\Users\Philip\AppData\Local\Temp\Engine_15412_46799beda7c646e097a18a06a953e544_\Engine_16788_e683c5b71e9b44f89eb069ef56b1fa7a_.yxdb
Single
Profile
C:\Users\Philip\AppData\Local\Temp\Engine_15412_46799beda7c646e097a18a06a953e544_\Engine_16788_7083629c91ba4a7fbabd24c888c1d0f6_.yxdb
Single
Report
Variable Importance Plot
Ten most important variables:
S_11, F_15, S_12, F_29, F_19, S_3, S_6, F_23, F_38, F_21
Horizontal
Challenge 157 - An Expert Challenge