数据指标波动假设检验统计

eda/recommended_indexs/hypothesis_test.py

@@ -181,24 +181,56 @@ def chi_DATA_yesterday(x,y,z,q,t1):
     CVR_DATA_yesterday = con_sql(sql_cid)[0]
     return CVR_DATA_yesterday
 
+#整理数据
+def data_cal(x,y):
+    x_a = [x[0], x[1] - x[0]]
+    y_a=[y[0], y[1] - y[0]]
+    a_df=pd.DataFrame({'原':x_a,'测':y_a})
+    return a_df
 chi_ctr_precise_recently=chi_DATA_recently("clk_count_oldUser_all_a","clk_count_oldUser_all_b","imp_count_oldUser_all_precise","on_click_diary_card",five_days,yesterday)
 temp1=[float(str(Decimal(chi_ctr_precise_recently[i]).quantize(Decimal('0.0')))) for i in range(len(chi_ctr_precise_recently))]
-temp1_a=[temp1[0],temp1[1]-temp1[0]]
-print(temp1)
+# print(temp1)
 chi_ctr_precise_yesterday=chi_DATA_yesterday("clk_count_oldUser_all_a","clk_count_oldUser_all_b","imp_count_oldUser_all_precise","on_click_diary_card",yesterday)
 temp2=[float(chi_ctr_precise_yesterday[i]) for i in range(len(chi_ctr_precise_yesterday))]
-temp2_a=[temp2[0],temp2[1]-temp2[0]]
-print(temp2)
+# print(temp2)
+tst=data_cal(temp1,temp2)
+print(tst)
 
-a3=pd.DataFrame({'原':temp1_a,'测':temp2_a})
 
-a3['共计'] = a3.apply(lambda x: x.sum(), axis=1)
-print(a3)
-a3.loc['共计'] = a3.apply(lambda x: x.sum())
-print(a3)
 # print(chi_ctr_precise_recently)
 # print(chi_ctr_precise_yesterday)
 
+def chi_cal(data):
+    data['共计'] = data.apply(lambda x: x.sum(), axis=1)
+    # print(data)
+    data.loc['共计'] = data.apply(lambda x: x.sum())
+    # print(data)
+    t1=data.iloc[0]
+    t2=data.iloc[1]
+    t11_count=t1[0]
+    t12_count=t1[1]
+    t21_count=t2[0]
+    t22_count=t2[1]
+
+    ###理论值计算
+    temp1=data.loc['共计']
+    rate1=temp1[0]/temp1[2]
+    rate2=temp1[1]/temp1[2]
+    temp2=data.iloc[2]
+    t11_theory=temp2[0]*rate1
+    t12_theory=temp2[1]*rate1
+    t21_theory = temp2[0] * rate2
+    t22_theory = temp2[1] * rate2
+    #计算卡方值
+    X=((t11_count-t11_theory)**2)/t11_theory+((t12_count-t12_theory)**2)/t12_theory+((t21_count-t21_theory)**2)/t21_theory+((t22_count-t22_theory)**2)/t22_theory
+    #计算自由度
+    v=(len(data)-1)*(data.columns.size-1)
+    #查表发现阈值为3.84
+    if X>3.84:
+        print("数据波动较大，超出正常波动范围，95%可能性属于指标显著变化，请关注")
+    else:
+        print("数据波动较小，95%可能性属于正常波动范围")
+
 # chi_cvr_new=
 # chi_cvr_old=
 #