1 00:00:00,930 --> 00:00:06,450 So now that we know the basic concept behind logistic regression, we'll be learning a simple model 2 00:00:07,080 --> 00:00:09,690 in which we will use only a single predictor. 3 00:00:10,560 --> 00:00:12,360 So let us look at this data. 4 00:00:15,810 --> 00:00:23,250 We want to predict the values of this all variable, whether the house will be sold within three months 5 00:00:23,250 --> 00:00:23,700 or not. 6 00:00:23,850 --> 00:00:27,030 If it is if it will be sold, it has value. 7 00:00:27,450 --> 00:00:29,570 And if it will not be sold, it has value, too. 8 00:00:31,590 --> 00:00:33,330 So we will be using only one predictor. 9 00:00:33,510 --> 00:00:34,530 Which is this price. 10 00:00:34,650 --> 00:00:36,600 That is the asking price. 11 00:00:37,500 --> 00:00:38,550 How much is the price? 12 00:00:38,640 --> 00:00:40,740 So that is asking for this property. 13 00:00:44,380 --> 00:00:49,650 So running a logistic model is very easy to create a logistic model. 14 00:00:50,190 --> 00:00:56,760 We need to use a function called DMM so we will make delimit outfit. 15 00:00:57,390 --> 00:01:01,380 This is the variable name so you can keep anybody of a name. 16 00:01:01,920 --> 00:01:05,190 But after that, you need to write any function, which is Dehlin. 17 00:01:09,240 --> 00:01:13,290 So this is a function which stands for generalized linear models. 18 00:01:14,700 --> 00:01:19,090 The first parameter is the response variable, which is sword. 19 00:01:19,480 --> 00:01:20,490 So Wheelwright sold. 20 00:01:25,000 --> 00:01:26,270 So this is the response very well. 21 00:01:26,830 --> 00:01:29,970 And it is to be dependent on one productivity Bill. 22 00:01:30,160 --> 00:01:35,300 So will they do this in whatever way they don't deliver contained or they're delivering? 23 00:01:35,530 --> 00:01:38,470 Is the dependent variable or the response variable? 24 00:01:38,710 --> 00:01:44,860 And whichever variable, Wilmington on the right hand side, that will be predictor variables. 25 00:01:45,820 --> 00:01:46,840 You're right, price. 26 00:01:49,630 --> 00:01:52,570 So these two variables we need to use to create this money. 27 00:01:53,380 --> 00:01:55,480 And we'll also specify the date that we used. 28 00:01:55,490 --> 00:01:57,070 We just do days equal to be if. 29 00:02:03,220 --> 00:02:07,510 The third parameter that we have to specify is family is equal to binomial. 30 00:02:08,620 --> 00:02:15,370 This barometer will tell us that we are running a logistic regression rather than any generalized linear 31 00:02:15,370 --> 00:02:15,700 model. 32 00:02:16,390 --> 00:02:18,940 So this family is equal to binomial is important. 33 00:02:18,940 --> 00:02:28,260 But every day when you are running a logistic model to run this, you can see that a new variable dealing 34 00:02:28,310 --> 00:02:33,700 four discrete did this variable has the information of this logistic model. 35 00:02:35,800 --> 00:02:38,560 Look at the information you can make somebody 36 00:02:41,430 --> 00:02:42,440 in the back roads. 37 00:02:42,460 --> 00:02:47,980 Will IDL and Broadford just be ready, willing and will run it? 38 00:02:53,200 --> 00:02:58,760 So this is all the information that we get when we run the logistic regression model. 39 00:03:00,300 --> 00:03:02,430 But we are going to discuss in the next lecture. 40 00:03:02,550 --> 00:03:10,440 Is this part in which we are getting the relationship of price with this old video, but which is the 41 00:03:10,440 --> 00:03:11,310 response variable? 42 00:03:12,540 --> 00:03:16,000 So we wanted to predict B does you don't beat 11 values. 43 00:03:16,980 --> 00:03:19,440 This estimates column is giving us this. 44 00:03:19,440 --> 00:03:19,830 B does. 45 00:03:19,830 --> 00:03:26,140 You know, we did the intercept and this beta one, which is decommissioned of price variable. 46 00:03:27,150 --> 00:03:32,820 So we have B does it when we do one in the next video, we will understand what is the meaning of all 47 00:03:32,820 --> 00:03:33,780 these other values.