WEBVTT 00:00.270 --> 00:01.020 Hello, everyone. 00:01.320 --> 00:04.170 So let's start working on a case study. 00:04.650 --> 00:11.610 And for this, we will be using a commonly used dataset, which is how how is price prediction? 00:12.240 --> 00:16.770 So let us import the basic libraries, which we will need. 00:20.760 --> 00:22.920 So we have important non-pay. 00:23.280 --> 00:24.300 US, you, my. 00:25.380 --> 00:33.360 And we have also important the data set which we will in needed for the latest, Jabe Schabel, the 00:33.360 --> 00:33.810 data. 00:33.810 --> 00:38.430 So we can simply say to you don't want to get the data sheet. 00:38.760 --> 00:43.800 So here you can see there are fourteen hundred and sixty rows and 81 columns, which we have. 00:44.370 --> 00:51.750 So let's have a look at the data first so we can safely thought or tend to view the data. 00:52.050 --> 01:01.020 But first, we can also set some properties in pandas so that we can view all the data points, at least 01:01.020 --> 01:02.040 all the columns. 01:15.060 --> 01:19.950 This will allow us to view all the columns and once we have said it to anyone. 01:20.400 --> 01:23.310 Now we will fund this. 01:23.970 --> 01:27.630 And you can see we are able to view all the data points at once. 01:27.990 --> 01:30.210 So what does the status quo? 01:30.270 --> 01:40.200 So this is a data regarding house prices and different parameters, which actually are present for estimating 01:40.200 --> 01:41.070 that house price. 01:41.460 --> 01:45.240 So this is the column sale price, which is the price of the house. 01:48.520 --> 01:52.500 Further, we have different properties regarding the house to be rented. 01:52.860 --> 01:58.380 This is simply an IED, which cannot which does not hold any specific information. 01:58.710 --> 02:01.320 It is just a unique identifier for each and every house. 02:01.980 --> 02:05.910 Then we have a subclass and a zoning. 02:05.940 --> 02:08.380 What is the log frontage? 02:08.430 --> 02:09.520 What is the area? 02:09.540 --> 02:10.680 What is the street? 02:11.760 --> 02:15.510 Which alleys on the sheep, Dequan? 02:15.510 --> 02:16.740 Door of the land? 02:17.130 --> 02:19.290 The utilities which are available? 02:19.410 --> 02:22.740 What is the neighborhood is what is the condition of the house? 02:23.040 --> 02:25.380 And several other details are present. 02:25.950 --> 02:33.570 You can also view delete a dictionary which is available with the data and using that. 02:33.570 --> 02:39.240 Additionally, you can easily find out what all the information is presented, this data. 02:39.690 --> 02:46.140 So usually what we do is we actually go through the entire data dictionary to understand what type of 02:46.140 --> 02:46.980 data we have. 02:47.400 --> 02:55.770 And for that, we look at the data frame itself and we try to identify what type of columns are present 02:56.250 --> 02:59.970 and different attributes of a particular dataset. 03:00.300 --> 03:08.410 So for that, letters, of course, have a look at the names of the data so we can simply say they, 03:08.490 --> 03:10.450 Don, don't them? 03:17.540 --> 03:19.630 And you can see the columns of the data. 03:20.630 --> 03:25.190 And for the letters, also have a look at different. 03:26.060 --> 03:33.510 You can see different type of information so we can say data to inform full view. 03:34.400 --> 03:35.810 What type of data is present? 03:35.810 --> 03:41.510 So you can see IDs, integer and subclasses and data that are fourteen point values also. 03:41.510 --> 03:44.660 And there are a lot of categorical columns which are present. 03:44.990 --> 03:46.760 You can see these are object. 03:47.390 --> 03:52.990 These are all categorical columns which have datatype as an object for there. 03:53.010 --> 03:55.740 You can also see the number of null values. 03:55.740 --> 04:00.780 So you can see that there are one four six zero, not null values here. 04:00.800 --> 04:04.490 You can see in the column there are a lot of null values present. 04:08.210 --> 04:11.840 Here again, in five place, there is a lot of null values present. 04:12.170 --> 04:15.650 So accordingly, you can decide which column you want to keep. 04:15.680 --> 04:23.600 Usually I will not keep any column which is having more than, let's say, 50 percent or more than 20 04:23.600 --> 04:25.280 percent, no data. 04:25.850 --> 04:27.860 So it's a big no. 04:27.890 --> 04:29.960 So we remove these columns. 04:31.340 --> 04:32.000 Flameless. 04:32.000 --> 04:38.120 Q You will see things, miscellaneous features. 04:38.390 --> 04:44.960 You can remove all of these because these will not really impact will not hold much information or do 04:44.990 --> 04:53.120 what you can do is you can have a view of the mean median more than important metrics of the numeric 04:53.120 --> 04:53.810 variables. 04:54.050 --> 04:59.480 So for that, what you can simply do is you can see a detente or that's great. 05:02.750 --> 05:08.540 And here you can see the count, the mean, the standard deviation, the minimum value, the maximum 05:08.540 --> 05:09.050 value. 05:09.380 --> 05:14.680 Then you can have a look at the twenty five percentile books and ten, seventy five, PopTech, whatever. 05:14.870 --> 05:18.050 Do they split and then you can accordingly have a look at that. 05:19.730 --> 05:27.750 You can see this seems to be a categorical value just Kadence. 05:28.070 --> 05:29.930 Here we have zero one, two, three. 05:29.940 --> 05:32.240 So maximum three bathrooms are present here. 05:32.750 --> 05:38.300 This is how but maximum to have but sufficient so that you have all the information which is present 05:38.300 --> 05:38.570 here. 05:38.900 --> 05:46.250 You can see the maximum price of our house is seven point five thousand. 05:46.490 --> 05:53.000 Well, if you see the minimum value, is this here? 05:53.660 --> 05:54.890 This is the minimum value. 05:56.870 --> 05:58.070 This is the minimum value. 05:58.490 --> 06:07.400 And the maximum value is very much different from the seventy five percent in value. 06:07.670 --> 06:12.380 So here you can easily see that the values are here. 06:12.390 --> 06:17.780 We have a lot of outliers, but we will be able to identify and analyze that later. 06:19.280 --> 06:20.570 Let's go further. 06:20.840 --> 06:32.000 And let's see the different sale prices and let us actually have a lot of the distribution of the sales 06:32.000 --> 06:32.540 prices. 06:34.870 --> 06:44.800 So for that, what we can do is we can simply say it's this thought, this song, which is the distribution 06:44.800 --> 06:48.930 block, and inside this, you can give data and in data. 06:48.930 --> 06:59.640 I want to have a look at the sales price column so I can put appeal and order. 07:00.940 --> 07:07.210 I can simply say not not sure. 07:08.190 --> 07:08.350 And 07:11.710 --> 07:17.480 now here you can easily see that this is the distribution. 07:17.500 --> 07:28.510 Initially, the data looks nicely distributed, but later you can see there are a lot of points which 07:28.510 --> 07:31.160 are lying towards the right thing. 07:31.480 --> 07:34.840 So here you can see that data is skewed in nature. 07:37.980 --> 07:41.710 For though, you can easily find out the mean median mode. 07:41.730 --> 07:48.660 Although we would we have seen the mean of these values, you can still find out the median, the mood, 07:50.040 --> 07:52.290 the skewness, the Cardoso's. 07:55.080 --> 07:56.610 So let's do that first. 07:59.160 --> 08:07.830 So here I have retrieved all the important metrics that is minimax mean, median Morde, standard deviation 08:07.830 --> 08:09.780 readings, cutlasses and skewness. 08:11.280 --> 08:14.700 So you can see it is highly lepta political protecting nature. 08:16.320 --> 08:17.580 It is highly skewed. 08:18.490 --> 08:21.630 So you can easily see these other data points, which we have. 08:22.110 --> 08:23.520 You see, this is the main value. 08:23.520 --> 08:24.870 This is the max value. 08:25.320 --> 08:26.760 This is the mean value. 08:27.420 --> 08:28.830 And this is the median value. 08:28.920 --> 08:32.760 You can easily see that the mean and the median value is almost similar in nature. 08:33.090 --> 08:38.820 So there is not much difference because here there are very few data points which are present in the 08:38.820 --> 08:39.180 team. 08:42.840 --> 08:45.030 So this is what we have got from this. 08:45.750 --> 08:53.670 Now, further, what we can do is we can try to find out the sample and population mean. 08:53.670 --> 08:57.730 Let's try to identify something from this particular data. 08:57.780 --> 09:02.690 Let's see if a sample will be similar to the population itself. 09:02.700 --> 09:04.440 So let's try to find that out. 09:06.240 --> 09:08.370 So for this, I can fix a. 09:08.710 --> 09:11.700 So let's say in Bidoun flying now. 09:12.780 --> 09:13.890 So it seemed. 09:14.850 --> 09:17.850 So here I am giving some seats, 23. 09:18.780 --> 09:22.460 And now let's take some sample values. 09:22.460 --> 09:25.660 So let's take oh, how many data points out there? 09:25.680 --> 09:29.220 So the shape is one four six zero. 09:29.240 --> 09:31.590 So let's take 500 data points from this. 09:34.020 --> 09:37.930 So we can simply see something. 09:45.600 --> 09:50.700 Each is equal to impede or random 09:55.580 --> 09:55.890 for. 10:12.740 --> 10:27.730 It's his price, and out of this, I want to get 500 values so I can see size is equal to 500. 10:28.500 --> 10:29.950 So this is my south village. 10:30.520 --> 10:39.040 And for this, I want to get the mean to what is to me with this, I can just simply pectus and find 10:39.040 --> 10:41.020 out the meaning of this. 10:42.400 --> 10:48.370 So here you can see that the sample price. 11:00.240 --> 11:03.510 It's nothing but one eight one six seven five. 11:04.170 --> 11:11.580 Now, if we look at the population mean we can basically find all the means we hear already we have 11:11.580 --> 11:11.740 it. 11:12.240 --> 11:14.700 So this will see something mean the third one. 11:15.270 --> 11:20.100 So this is the population mean and this is the sample mean. 11:20.100 --> 11:23.310 So you can see these are clearly very close to each other. 11:25.350 --> 11:32.490 Now, next, what we will be doing is we will try to find out a little more about it. 11:34.410 --> 11:37.470 Now, let us have a look at the few neighborhoods which are present. 11:39.090 --> 11:44.720 So for that, we can have a look at these numbers, the work. 11:49.070 --> 11:50.390 These are the names. 12:01.520 --> 12:04.520 So these are the different neighborhoods which are present. 12:05.000 --> 12:18.560 Now let's see for Edward if the prices in Edwards is similar to the prices in the entire 12:21.350 --> 12:21.840 place. 12:21.840 --> 12:31.100 So what we do is we can simply try to find the comparison between both of them and we can compare this. 12:33.290 --> 12:39.470 So for this, we will be performing the sectors because we are comparing a sample to the population 12:40.220 --> 12:43.610 and we want to find out if the sample is different from the population. 12:44.240 --> 12:49.880 We want to compare a particular neighborhood to the entire population, which we have. 12:49.910 --> 12:50.750 So let's do that. 12:51.110 --> 13:01.580 Before that, we will first need to import the library for the test so we can simply says from. 13:08.120 --> 13:11.260 From stats Mondelēz about the stats. 13:17.120 --> 13:17.990 With stand. 13:24.750 --> 13:33.290 Because it is so wildly important, said that we will be using this particular test to find out if the 13:33.860 --> 13:39.110 population is similar to the neighborhood or not, so we can save said. 13:39.980 --> 13:53.900 Well, and you do have that as an output from this test and say that this we will have to pass the population 13:53.900 --> 14:03.830 as well as the sample, so we can simply say X one is equal to data. 14:06.700 --> 14:06.950 They 14:13.540 --> 14:18.740 would be done a very good is equivalent to it punch. 14:24.290 --> 14:31.970 Now, from this, I want to get these sales price so and sees his price hill. 14:44.290 --> 14:51.850 Now, apart from this, I need to provide the entirety of space for the population so I can simply say 14:52.660 --> 15:00.680 values is equal to Peter still space. 15:02.410 --> 15:08.170 Let us again have a look at the documentation to be sure of it so we can see. 15:14.480 --> 15:17.600 His leftist. 15:29.400 --> 15:31.590 So you can see it takes value 15:35.160 --> 15:42.360 and the value has to be they mean, so we'll see. 15:42.840 --> 15:44.090 I mean. 15:49.920 --> 15:56.760 Now let's run this and also let us bring these Etoile and Biegun. 16:17.290 --> 16:28.240 So you can see that does did well is minus 12 and the P-value is very small, which means that the neighborhood, 16:28.690 --> 16:33.580 Edward, is significantly different from the other neighborhoods. 16:34.630 --> 16:41.440 Now, let us perform the same thing for another neighborhood so we can fix the entire thing from here 16:42.730 --> 16:45.910 and we can replace it with another neighborhood name. 16:49.870 --> 16:52.900 Let's take Boystown for now. 17:06.560 --> 17:08.280 And it's this. 17:09.180 --> 17:12.180 So here again, the value is very small. 17:13.440 --> 17:27.000 So you can easily see that the mean for the location, Edward, and the entire population is different. 17:27.420 --> 17:34.350 And similarly, the mean for the location or town is also significantly different from the entire population. 17:35.430 --> 17:38.340 So this is the data which we have. 17:38.340 --> 17:43.680 So we are just trying to see that the means would be different for both for the sample that we have 17:43.680 --> 17:46.740 picked for a particular neighborhood and for the entire population. 17:49.620 --> 17:56.490 Now, when we look at this data, we can easily see that the neighborhood is having a very important 17:56.490 --> 17:58.260 impact on the teams faces. 17:58.590 --> 18:03.570 So if there is one neighborhood, there would be different type of sales places, which would be pleasant 18:03.780 --> 18:04.800 in another neighborhood. 18:04.800 --> 18:08.610 There would be some other type of sales places would be which would be pleasant. 18:09.060 --> 18:13.470 Now, for the weekend here, we have compared to the population. 18:13.770 --> 18:20.630 We can also compare the city's prices and different factors amongst the neighborhoods. 18:20.640 --> 18:25.130 Is that because these will be two different detours? 18:25.210 --> 18:29.700 The one would be, for example, neighborhood and there would be for Vye Neighborhood. 18:30.090 --> 18:33.300 So here we are comparing two different samples. 18:33.810 --> 18:39.600 So based on two different samples, we can try to find out if those two samples are somewhat similar 18:39.600 --> 18:40.680 to each other or not. 18:41.130 --> 18:43.020 But that is also another thing which we can do. 18:43.320 --> 18:49.320 But whenever we are dealing with two different samples and not comparing with the population, that 18:49.320 --> 18:55.530 is the time when the other tests come into which that is like the disguise criticized. 18:55.800 --> 18:57.450 Those tests will come into picture. 18:57.750 --> 19:04.410 When we are talking about population, we will try to solve it using the Zetas because their test gives 19:04.790 --> 19:06.630 a population into consideration. 19:08.370 --> 19:15.060 Now, let's try to identify how many houses would I be able to get if I had a certain amount of money. 19:16.230 --> 19:24.510 So let's first find out the details about the neighborhood input. 19:28.520 --> 19:36.140 So here, let me get the data for Atwood and for the same. 19:37.190 --> 19:41.570 I will find out what the mean value and standard deviation. 19:42.380 --> 19:43.340 So the means 19:46.970 --> 19:47.480 to me 19:53.420 --> 19:58.250 and it would be the standard deviation. 19:59.160 --> 19:59.360 OK. 20:02.690 --> 20:07.550 Now, let us import that package from 20:11.650 --> 20:12.400 before. 20:15.050 --> 20:22.130 That's usually what we do is we imake on imports at the top of the file so that if someone else is looking 20:22.130 --> 20:25.580 at our code, they'll be able to find out more likely to be imported. 20:25.880 --> 20:30.060 So usually in take this example of your coffee. 20:30.620 --> 20:34.360 So that is find out the Z score. 20:34.370 --> 20:41.990 So here, as we know, that code is equal to the value. 20:42.170 --> 20:49.160 So let's say I want to find out how many houses I will be able to get if I see something that's 50000. 20:50.760 --> 20:53.120 Then I looked the mean from it 20:56.600 --> 21:00.920 and I divide it from the standard deviation. 21:02.720 --> 21:06.710 And I want to find out later the percentage. 21:07.310 --> 21:17.480 So the percentage can be from Goldman Sachs thought, ma'am, thought C, the F CVS will give you the 21:17.960 --> 21:23.350 probability, the percentage probability from the Z score. 21:23.840 --> 21:25.700 So we put into that scoring in. 21:33.170 --> 21:40.870 So here there is ninety nine point seven, five percent probability that you will get a house in Everwood 21:40.890 --> 21:43.640 if you are paying 250 thousand. 21:44.540 --> 21:47.540 Now let's decrease it a little bit. 21:47.540 --> 21:49.400 So let's say sort of 250 thousand. 21:49.400 --> 21:51.650 I'm willing to pay only 180 thousand. 21:52.250 --> 21:54.860 So let's see how much is the probability. 21:55.100 --> 22:02.210 So there is a B eight point four, six percentage that I will be able to find a good house for myself. 22:03.390 --> 22:09.620 Now, let's do the same and check for some other neighborhood. 22:10.130 --> 22:12.380 So let's have a look at. 22:13.640 --> 22:14.030 So I 22:19.220 --> 22:27.110 see there are very few houses there, so let's check what is to start with that particular place. 22:27.680 --> 22:33.200 So let's first check for 250 and drive this. 22:33.920 --> 22:36.510 So here you can see that is eighty seven point two. 22:36.530 --> 22:37.390 It was intense. 22:37.750 --> 22:40.320 And I'll get a house in science. 22:40.730 --> 22:44.420 So this means that the amount would have been higher. 22:44.420 --> 22:45.570 The range would have been higher. 22:45.590 --> 22:49.670 So let's try to look at the mean for this particular location. 22:50.330 --> 22:58.850 So, I mean, is a one eight six triple five that get the maximum value? 23:10.200 --> 23:12.620 See, here is a higher value. 23:12.920 --> 23:14.330 So let's put. 23:17.510 --> 23:25.550 So even after paying a much higher amount in comparison to input, there is a comparatively lower probability 23:25.550 --> 23:27.200 that I'll be able to get a house here. 23:27.560 --> 23:38.630 So as we see Ed Wood and Soyer have comparatively much difference in the max value and the values which 23:38.630 --> 23:47.000 I'm getting into that school, let's try to find out from the samples of both if these would be actually 23:47.000 --> 23:51.350 different from each other or are they are similar in nature. 23:51.560 --> 23:59.410 So if I take a sample from somewhere and if I did a sample from Edwards, let's try to find out if the 23:59.690 --> 24:03.970 those prices are significantly different in Edwards. 24:03.970 --> 24:04.010 And. 24:04.420 --> 24:05.780 So let's let's try that. 24:10.220 --> 24:16.130 So for that, first, we need to get the data for both the locations. 24:16.520 --> 24:23.410 So let's get it inside a piece. 24:24.770 --> 24:30.050 So we say E is equal to the data for soy 24:39.620 --> 24:44.940 and the beans equal to the data for October. 24:52.340 --> 24:57.450 Now, let's do a thesis on this independently so that. 24:58.560 --> 25:03.870 So the the other schools before independent. 25:04.640 --> 25:06.620 And we have kids in here. 25:07.250 --> 25:13.630 And we give amethysts zero and equal. 25:13.750 --> 25:16.010 Explain that. 25:18.200 --> 25:19.040 And it's from this. 25:21.830 --> 25:24.620 You can see that the P value is very low. 25:25.070 --> 25:31.370 Hence, we can see that these the prices in Soyer and in gold are significantly different. 25:31.790 --> 25:42.350 Now, let's actually try to find out that with how much amount I will be able to be 90 percent sure 25:42.350 --> 25:51.650 that I will get a house in, let's say, in Englewood or or down or any other location. 25:51.650 --> 25:53.150 So let's try to find that out. 25:53.810 --> 25:56.330 So let's do it below this one also. 25:58.400 --> 26:03.560 So, yes, let's use these same means and standard deviations. 26:04.460 --> 26:06.650 So now let's check foreswear. 26:06.920 --> 26:11.810 So let's see how in how much amount in what interval. 26:12.050 --> 26:16.370 I will be almost 90 percent sure to find out that I be able to get their house. 26:16.370 --> 26:16.490 And. 26:17.600 --> 26:27.950 So for that, I can simply do with the fact that they don't know what I mean. 26:27.960 --> 26:31.240 So I guess we need to provide how much was did we want to show it with? 26:31.350 --> 26:34.110 How much probability we want to be sure with. 26:34.530 --> 26:35.910 So let's say 90. 26:38.970 --> 26:41.710 Then we need to provide the details. 26:42.240 --> 26:46.610 So let's say we're looking for societies. 26:46.830 --> 26:49.660 So let's get new places for societies. 26:52.020 --> 26:53.940 Then we need to provide the means. 26:56.850 --> 27:00.390 And the skills would be the standard deviation. 27:03.150 --> 27:08.120 Now, we can simply run this. 27:08.190 --> 27:13.650 And these are the different values which we are getting. 27:17.720 --> 27:21.860 We have to provide the lens here, so we say, and then 27:26.450 --> 27:29.720 this will give us the DNA P-value. 27:30.200 --> 27:33.080 So here. 27:36.570 --> 27:40.180 You can see the blower and a button below. 27:40.830 --> 27:48.540 So we need to have at least two, seven, nine typifying to be sure that there is a 90 percent probability 27:48.540 --> 27:51.030 of getting them or getting the house here. 27:51.300 --> 27:53.520 So let's look at some other 27:56.580 --> 27:57.440 neighborhood. 27:57.610 --> 28:00.810 So let's check for Edwards. 28:01.260 --> 28:02.670 Let's run this entire thing. 28:02.730 --> 28:03.720 Put it Curtice. 28:11.110 --> 28:13.360 So we need to get assistance. 28:14.550 --> 28:24.580 The experts that strung this post so far, 90 percent probability, we need to have around 28:27.130 --> 28:28.300 200000. 28:29.590 --> 28:36.730 Let's get further and let's try to look at it so at some other way. 28:37.300 --> 28:50.050 So we actually saw that when we are comparing the neighborhood for approval for the actual sale price 28:50.050 --> 28:53.920 of the entire data, it is significantly different. 28:54.550 --> 28:57.590 Same is the case of Georgetown. 28:57.970 --> 29:00.320 So for both, it is significantly different. 29:00.340 --> 29:07.010 And we found that using these mechanisms now is at best is usually used when we are comparing with the 29:07.510 --> 29:09.760 of the population values. 29:10.000 --> 29:13.540 And then we don't really have these samples in hand. 29:13.870 --> 29:20.530 But when we're dealing with samples, then in that case, in sort of using this as a test, we should 29:20.530 --> 29:22.680 have used even to repeat this. 29:23.050 --> 29:25.360 So let's apply the same. 29:25.360 --> 29:33.760 Let's try to find out if the neighborhood sample for neighborhood Edwards or the neighbor down or it 29:33.940 --> 29:38.850 is significantly different from the entire population means. 29:39.070 --> 29:40.630 So let's try to do that. 29:40.630 --> 29:42.190 Let's try to find that out. 29:46.420 --> 29:49.720 So let's take this from here. 29:51.310 --> 29:54.100 Let's get this news. 29:57.070 --> 30:00.060 So now we are applying the one somebody asked. 30:07.180 --> 30:12.550 So let's get the scores and the bailout deal. 30:14.230 --> 30:27.490 And this we will be getting from these steps are the best ideas for one step, not just for one sample. 30:28.300 --> 30:30.040 Then we get the data. 30:30.040 --> 30:34.270 So the data would be just the same from here. 30:36.790 --> 30:38.680 Let's check for a good. 30:42.260 --> 30:45.140 And this will be the population us. 30:53.650 --> 30:55.780 Now, this is people me. 30:59.540 --> 31:05.630 And here we have the data, so instead of providing the entire population, we'll have to provide the 31:05.660 --> 31:06.350 sample here. 31:06.740 --> 31:16.220 So let's take a sample of, say, 60 values, taking a sample of 60 values. 31:17.000 --> 31:17.630 Now 31:20.150 --> 31:24.800 let's get the Peace Corps and the P-value. 31:46.550 --> 31:51.590 So here you can see again, the P value is significantly low. 31:51.980 --> 31:58.670 Hence, we can easily see the same thing, that the neighborhood advert is significantly different from 31:58.670 --> 32:01.940 the population on the basis of one sample fetus. 32:02.240 --> 32:07.560 Same thing we can do repetitively for different neighborhoods. 32:07.940 --> 32:10.370 Let's pick a different neighborhood. 32:10.370 --> 32:12.300 So let's see. 32:13.730 --> 32:22.630 For now, for proper iliakis this and insert over 60 Nesquik just to be here. 32:23.600 --> 32:24.710 What is the actual. 32:25.490 --> 32:27.260 That's fifty one so they can be put to you. 32:28.370 --> 32:35.390 So here you can see here the P value is zero point zero two five three. 32:35.720 --> 32:39.920 So this is not much different. 32:40.400 --> 32:44.300 But yes, it is marginally at the exact point. 32:44.540 --> 32:47.780 So it completely depend on the P value, which you are looking at. 32:48.080 --> 32:54.080 So if you're looking at, let's say, one percent, then it is significantly different. 32:54.380 --> 32:59.930 If you are looking at 2.5 or 2.5 percent, then it is almost that. 33:00.200 --> 33:03.480 So this is completely up to you how you look at it. 33:03.510 --> 33:09.740 And again, if you're looking at, let's say, five percent P value, then it is way below the five 33:09.740 --> 33:12.860 percent above the five percent P value. 33:12.860 --> 33:15.290 Then you cannot see that it is significantly different. 33:15.620 --> 33:20.750 So that's how you can easily evaluate using the one sample doing that. 33:20.990 --> 33:24.650 So it completely depends on the significance level that you're looking at. 33:26.030 --> 33:31.880 Now, let's look at this particular location. 33:33.650 --> 33:36.600 You can do your own analysis. 33:36.620 --> 33:42.020 I'm just giving you an idea of how you would perform an analysis, how you would do that even in sight. 33:42.410 --> 33:46.190 Now, let's take an example of, say, 60 years. 33:47.810 --> 33:54.290 Here you can see this is very clearly visible, that that is a significant difference between the amount 33:54.290 --> 33:55.730 of opposition and quality growth. 33:56.030 --> 33:58.520 So that's how you get an insight photofit. 33:59.090 --> 34:03.260 Now, for there, we can have look at different now. 34:04.040 --> 34:10.010 We have been comparing a particular neighborhood with the entire population. 34:10.310 --> 34:18.050 Now, in case we want to compare one neighborhood with another neighborhood, here, you can see this 34:18.140 --> 34:23.570 is how we do it from the independent Betis to compare to different neighborhoods. 34:23.900 --> 34:29.780 So this is only the which we have under the best umbrella. 34:30.230 --> 34:33.800 We have the B, the independent the best. 34:33.800 --> 34:36.380 We have the one sample, the test. 34:36.650 --> 34:39.740 So you can perform these features like this. 34:40.070 --> 34:43.020 Then apart from that, we will have prepared people. 34:43.040 --> 34:44.810 So then we have the bare details. 34:44.840 --> 34:51.980 It has to be done when we have some particular data, which is the same sample is being tested for in 34:51.980 --> 34:53.030 another time frame. 34:53.390 --> 35:00.980 So in that case, in this particular data, we don't have that example, but you can surely use that 35:00.980 --> 35:03.450 in any other case from any other dataset. 35:03.460 --> 35:04.970 So that is completely up to you. 35:06.560 --> 35:12.740 But just to give you a particular insight, let me try to give you an example from this particular data. 35:13.190 --> 35:18.330 So so let's say be down and get the value for each day. 35:25.250 --> 35:27.730 Now, the t the values in it. 35:27.770 --> 35:29.420 So these are the different values. 35:29.780 --> 35:32.960 Now, let me plot these for you in the same plot. 35:35.540 --> 35:39.350 So this might be a little time to generate the plot. 35:39.680 --> 35:43.000 So here you can see the plot for the year boot. 35:43.400 --> 35:48.740 Now, when you see this, you can see that there are few houses which were built earlier and major lead 35:48.740 --> 35:51.920 in on houses which are very neat here. 35:52.460 --> 35:57.010 So let's try to compare the houses which were built. 35:57.780 --> 36:01.500 Nineteen sixty and the houses which have occurred after nineteen sixty. 36:02.020 --> 36:07.550 So let's do that and compare using the independent dataset. 36:08.190 --> 36:13.530 So let's see A and B and let it be. 36:14.560 --> 36:16.590 Now let's get this condition. 36:16.950 --> 36:17.790 So. 36:26.700 --> 36:29.120 Let's get just as much from 36:32.550 --> 36:34.980 now, we need to compare the year before, 36:39.120 --> 36:50.130 and let's say we're looking for a less than equal to 1960, the same space for that. 36:51.120 --> 36:57.480 And the ones which are based off the 1960s included in 1960. 36:58.230 --> 37:00.600 These are the two which we wanted to compare. 37:00.930 --> 37:11.850 Now, for the next to the datasource that don't be based on the school that might be four independent. 37:12.570 --> 37:28.320 And we provide in the and we see this equal to zero and b C for is equal to two. 37:29.700 --> 37:31.140 Now let's run this. 37:32.970 --> 37:36.320 So here you can see the P value is very, very, very low. 37:36.330 --> 37:43.620 So which means that there is a very highly significant difference between the sales prices of the housebreak 37:43.950 --> 37:45.510 before and after 1960. 37:45.550 --> 37:53.850 Now, let's do the same comparison on the houses, which are right after, let's see, 1980. 37:55.990 --> 37:58.350 Let's compare on the basis of 1980. 37:58.980 --> 38:04.200 So here again, the prices are very different. 38:04.860 --> 38:07.500 Now, let us strive for 1980. 38:18.190 --> 38:18.910 Same thing. 38:19.300 --> 38:25.830 So here you can see the result now. 38:29.260 --> 38:41.860 Let us move to the Chi Square umbrella test, so let us check if dolus different line slopes are independent, 38:42.250 --> 38:51.670 so let us perform a test to check if knowing a particular line condole will have some impact. 38:52.150 --> 38:57.970 So for that, we will have to divide the science place into different criteria. 38:57.970 --> 39:04.600 So we'll go divide those whose plays into high, medium and luteal spaces, and then we will compare 39:04.600 --> 39:07.450 the different line contours with each other. 39:07.780 --> 39:09.370 So let's do that. 39:09.640 --> 39:14.050 So for that, let me simply create. 39:14.260 --> 39:20.890 Let me, first of all, import the stats by we already had these tags. 39:21.250 --> 39:22.780 So that is perfectly fine. 39:23.200 --> 39:35.020 So let us write a very small function, which will basically compute the frequency for guys square test. 39:35.260 --> 39:40.360 And then we can perform Chi Square test using those frequencies. 39:40.660 --> 39:46.570 So let's just write the function and just write that function for you and then explain it to you. 39:48.760 --> 39:51.730 So here is the function that is very simple in nature. 39:52.060 --> 39:56.050 So I have defined this function, compute frequency, chi square. 39:56.380 --> 40:01.030 So what it does is it will simply take the two columns which you are trying to compare. 40:01.360 --> 40:05.530 So I am providing the X and Y here. 40:05.950 --> 40:10.420 So in actual, we are comparing willAnd, Gondor and basis. 40:11.080 --> 40:13.510 So now let's get this. 40:13.810 --> 40:21.610 Now what we do is we just create the crosstabs of these frequencies. 40:22.090 --> 40:30.070 And using that particular crosstabs, we will simply apply the guys who are best for contingency on 40:30.070 --> 40:30.730 top of it. 40:31.180 --> 40:32.560 I'm using these guys. 40:32.580 --> 40:33.850 What is the contingency? 40:33.860 --> 40:41.110 We will be able to find out if Limed Gondwe actually has an impact on this, use phrases or not. 40:42.100 --> 40:45.040 So let's simply run this now. 40:48.490 --> 40:54.670 So here I have done this so you can see it simply created the cross tab. 40:54.670 --> 40:57.340 And then I simply printed be crossed out for you. 40:57.700 --> 40:59.680 This is the cross tab which was generated. 40:59.680 --> 41:06.700 So you can see for our line one door, there are four line Condor's being key achiness, low level. 41:07.090 --> 41:12.830 And then there seems places divided into high, medium and low using the feed. 41:13.300 --> 41:21.940 You could you could basically isor function, which cuts the data or you can divide the data into equal 41:21.940 --> 41:22.450 quarters. 41:22.490 --> 41:30.160 So when I say three, when I see I've got this in the freezer, it creates three equal regions for all 41:30.160 --> 41:36.970 three equally distributed categories for me, based on the numerical values presented the basis. 41:38.170 --> 41:42.070 So here I have the guys with this to start testing. 41:42.070 --> 41:46.360 So it comes out to be twenty six and the P value comes out to be a very low value. 41:46.630 --> 41:48.040 It is almost zero here. 41:48.280 --> 41:54.340 So this shows that there is a very significant difference between the scenes faces with respect to the 41:54.340 --> 41:57.550 front line contours and their independent dimension. 41:58.030 --> 42:00.730 So this is what we get you. 42:00.880 --> 42:09.100 We get to know that there is a lot of independents in the land, on doors and their corresponding sets, 42:09.100 --> 42:11.650 prices for this. 42:11.710 --> 42:17.920 If we look at this particular P value, we get to know that when you're performing decries, grittiest, 42:18.310 --> 42:20.980 and we have this very small P value. 42:21.250 --> 42:26.530 So it also appears the relationship between the land control and business price. 42:26.860 --> 42:32.430 Now, when you see the land, Korngold is having a very high impact on the scenes places. 42:32.860 --> 42:33.220 Right. 42:33.370 --> 42:40.090 When the not the line control is level, the sales prices are comparatively higher. 42:40.420 --> 42:42.370 When the land condor is low. 42:42.550 --> 42:44.890 You can see the sales prices are very low. 42:45.490 --> 42:54.300 So this shows that there is a very important dependency of sales prices on the land Gondo. 42:54.550 --> 43:03.280 So although Dilan condos internally are independent in nature, the sales prices are highly dependent 43:03.280 --> 43:04.870 on the long line control. 43:05.140 --> 43:10.140 So basically, non control is a very important feature to identify the sales places. 43:10.410 --> 43:16.600 This is this is what you will find out using the guys with this, that if the sales price is dependent 43:16.600 --> 43:18.130 on the line or not. 43:18.400 --> 43:27.430 So when I say that the P value is very low, that means that the sales price is dependent on the line 43:27.430 --> 43:28.060 control. 43:30.620 --> 43:34.420 Very high now. 43:34.550 --> 43:44.480 It simply shows that the two variables, sales price and loan line control, they are not independent. 43:44.720 --> 43:50.510 They are highly dependent because price is very highly dependent on the line. 43:50.510 --> 43:51.070 Quantock. 43:53.780 --> 43:58.010 Now, let us perform the one the analysis for radians, which is an over. 43:58.340 --> 44:05.000 Now it is the user domain, whether there are any statistically significant differences between the 44:05.000 --> 44:08.270 means of three or more independent or unpredicted groups. 44:08.720 --> 44:11.720 So Raegan, let check for the same thing. 44:11.720 --> 44:13.550 Will check for Dilan Condole. 44:13.880 --> 44:18.620 So Dilan Condor's bank is low and level. 44:18.950 --> 44:21.440 They are significantly different from each other. 44:21.470 --> 44:23.240 There are different categories. 44:23.510 --> 44:25.790 So these are independent and related groups. 44:25.790 --> 44:32.570 So we can actually check if the sales prices for these are different internally or not. 44:32.960 --> 44:41.390 So we did see this seem to be able to see that the see they had an impact on two places. 44:41.390 --> 44:43.700 Significant difference expectancy's prices. 44:44.660 --> 44:46.760 But we were not able to compare them. 44:47.270 --> 44:49.400 So let's compare them internally themselves. 44:51.290 --> 44:53.330 So let's check again. 44:53.630 --> 45:00.440 Now, here, each group which we will be taking considering has to be having a different participant 45:00.440 --> 45:02.240 and each group should have a different take. 45:02.810 --> 45:09.800 So when we are checking at this year's prices, we are looking at two prices of these different locations. 45:09.800 --> 45:12.950 So their houses will be different for each location. 45:13.250 --> 45:14.750 So the condition is perfect. 45:15.350 --> 45:20.460 Now, next, we will be checking out basically what we are trying to find out. 45:20.640 --> 45:28.280 It is when we perform the one we know, what is the one win or best will actually tell us if these are 45:28.280 --> 45:29.180 different or not. 45:29.600 --> 45:36.980 So we will be comparing multiple categories here and finding out which one is actually significantly 45:36.980 --> 45:40.520 different from all of the categories or or less similar or not. 45:41.090 --> 45:48.800 So if even one of the the category is significantly different, there will be a significant value which 45:48.800 --> 45:49.550 will be coming out. 45:49.910 --> 45:53.780 So first, let's applaud the group for this. 45:54.080 --> 45:56.960 And we also applaud the book's plot for the theme. 45:57.260 --> 46:01.910 I'm grouping the data by Land Cawdor and getting the sales places. 46:02.240 --> 46:06.070 And later, imploringly land control over this is prices. 46:06.080 --> 46:07.190 So let's try this. 46:07.520 --> 46:14.300 So here you can see I'm getting the land condors and their respective count mean, standard deviation 46:14.300 --> 46:15.440 and the maximum value. 46:15.770 --> 46:19.930 There you can see the mean for bank is one four three one zero four here. 46:19.970 --> 46:22.040 Twenty three, twenty nine, eighteen. 46:22.510 --> 46:29.510 Here the maximum values are not much different here, but for level the max value is quite high. 46:29.870 --> 46:39.960 Those 75 percent beilis somewhat similar for all for 50 percent nihil is again higher for the low one. 46:39.980 --> 46:42.740 So these are the details which we are giving from these. 46:43.040 --> 46:48.260 And total data is also different for bank we have for bank hills. 46:48.260 --> 46:52.070 And Lord, we have smaller number of houses. 46:52.310 --> 46:54.890 Microlevel, we have higher number of houses. 46:56.300 --> 46:59.150 So now let's look further. 46:59.510 --> 47:03.770 So we'll run the slides for if one way ANOVA. 47:04.070 --> 47:06.680 So here when we run this, we are checking. 47:06.950 --> 47:07.820 We are running it. 47:07.820 --> 47:16.340 We are comparing it for these places of land, condo level, for fighting for the low and for height. 47:16.820 --> 47:18.200 Now, let's run this. 47:18.920 --> 47:25.250 So when we run this, we get the value as a very high, very low value, which means that at least one 47:25.250 --> 47:29.570 of the classes is significantly different from all under three. 47:29.870 --> 47:31.880 So at least one of them is different. 47:32.150 --> 47:37.900 So what you can do is you can compare these you can compare these with one another and then find out 47:38.210 --> 47:40.370 what I even further does here. 47:40.380 --> 47:43.760 And you can see that be countered slightly different. 47:44.030 --> 47:49.310 So let me take a sample of B five from each one of these. 47:50.090 --> 47:53.420 So let's take a sample of thirty five from each one of these. 47:54.290 --> 47:56.180 So let's get this again. 47:59.000 --> 48:02.870 And they could go something 48:06.870 --> 48:07.460 like this. 48:08.740 --> 48:12.530 They've done this for. 48:29.420 --> 48:36.890 Here you see still the P-value is coming out to be very low, which simply means that there is at least 48:36.890 --> 48:42.440 one category which is having a significant difference from all of those three categories. 48:42.740 --> 48:46.160 Now, this actually calls for the further analysis. 48:46.460 --> 48:53.150 So what you will be doing is really comparing level with bank level with low level the 10th. 48:53.450 --> 48:58.490 And then you will find out which one is actually different so that you can get an idea what you will 48:58.490 --> 48:58.880 be doing. 48:59.150 --> 49:01.340 And you can simply do that using this. 49:01.850 --> 49:04.500 So that does not involve you can easily find that out. 49:07.040 --> 49:09.230 So this is one Vilanova. 49:09.470 --> 49:16.310 And similarly, you can try this out for much different multiple variables. 49:16.310 --> 49:18.620 So I have considered like one good. 49:18.620 --> 49:23.630 You can pick up any of this activity and then walk on it and then test.