1 00:00:00,670 --> 00:00:05,550 I'm glad that you're here and I'm glad that you have made it this far in the cause. 2 00:00:05,590 --> 00:00:09,260 And even if you skipped some sections that's not a problem. 3 00:00:09,300 --> 00:00:14,620 I understand that you may just be interested on doing some stock market analysis. 4 00:00:14,940 --> 00:00:16,080 So that's fine. 5 00:00:16,190 --> 00:00:21,340 But be sure that you know about Pandas and the Bokeh library. 6 00:00:21,660 --> 00:00:28,500 So we use pandas to handle data or whether it is stock market data or whether data or anything we use 7 00:00:28,500 --> 00:00:32,990 that for handling data in Python, analyzing, and grouping and etc. 8 00:00:33,630 --> 00:00:37,660 And also we use a balky library which is a great library for visualizing data. 9 00:00:37,860 --> 00:00:39,970 So that was an intro for this lecture. 10 00:00:40,080 --> 00:00:46,650 Now what you'll get from this lecture is you'll learn how to get stock market data and also other kinds of 11 00:00:46,650 --> 00:00:47,410 data. 12 00:00:47,970 --> 00:00:54,930 And the good thing of this is that you will not download these data from a web site or from your browser, 13 00:00:55,440 --> 00:01:03,790 but you'll get them directly from Python, and the way you do that is by using a library called pandas_datareader. 14 00:01:04,320 --> 00:01:16,170 So you need to install that library and you can install it with pip so install pandas underscore 15 00:01:16,320 --> 00:01:18,410 datareader. 16 00:01:19,410 --> 00:01:26,310 So this used to be actually a module under the Pdas library, but now it's a stand alone library. 17 00:01:26,320 --> 00:01:29,450 so this is how you install it. 18 00:01:30,630 --> 00:01:35,670 I have installed it already so it's says requirements already satisfied, but it shouldn't be a problem 19 00:01:35,670 --> 00:01:37,260 to install it. 20 00:01:37,350 --> 00:01:44,940 So once you do that you can use and of course I suggest you to use the Jupiter Notebook. 21 00:01:46,170 --> 00:01:53,160 So I was in this folder and now if I create a Jupiter file here. 22 00:01:54,450 --> 00:02:00,030 So let's call this stock analysis. 23 00:02:00,890 --> 00:02:02,720 Now I should have that file in here. 24 00:02:02,730 --> 00:02:06,180 So this is the notebook that I created. 25 00:02:06,240 --> 00:02:07,480 Great. 26 00:02:07,500 --> 00:02:12,610 So how do you go about getting these data in Python without even opening the browser? 27 00:02:12,660 --> 00:02:16,830 Well, that's actually fairly easy. 28 00:02:16,830 --> 00:02:21,870 For instance let's try to get recent data from Yahoo. 29 00:02:21,930 --> 00:02:26,000 For some companies that sell stock in there. 30 00:02:26,430 --> 00:02:34,890 So let's say Apple and the first thing you want to do is you want to import the required tool from the 31 00:02:34,890 --> 00:02:42,210 pandas_datareader library and one tool that is used to do to get the stock market data is the 32 00:02:42,210 --> 00:02:42,840 data tool. 33 00:02:42,930 --> 00:02:53,280 So from pandas_datareader import data .Now once you do that ATL 34 00:02:53,400 --> 00:02:57,140 Enter to execute the line, and once you do that. 35 00:02:57,150 --> 00:03:02,900 you want to use a method called DataReader. 36 00:03:03,780 --> 00:03:05,190 And if you ask for help. 37 00:03:05,190 --> 00:03:08,340 So just a question mark there, execute. 38 00:03:08,930 --> 00:03:14,900 You'll see the docstring, so the definition of this function. 39 00:03:15,000 --> 00:03:21,240 So this imports data from a number of online resources and this library can get data from Yahoo 40 00:03:21,240 --> 00:03:28,790 Finance, and Google Finance, and St. Louis FED or FRED and Kenneth French's data libraries. 41 00:03:28,800 --> 00:03:34,750 So and you can get data from probably all the existing companies out there. 42 00:03:35,050 --> 00:03:44,400 And now, to get the data from these resources, and from different companies what you can do is you want to 43 00:03:44,400 --> 00:03:47,570 pass some parameters to this data_reader function. 44 00:03:47,700 --> 00:03:54,240 Now the first parameter you may want to input there is the name parameter which expects the name 45 00:03:54,240 --> 00:03:54,930 of the company. 46 00:03:54,930 --> 00:04:00,720 Actually it's not the name of the company, of the company but it's a stock symbol. 47 00:04:00,810 --> 00:04:12,510 So stock symbols. If you search for stock symbols and you look at this list of symbols from the New York 48 00:04:12,810 --> 00:04:15,100 Stock Exchange, you'll see 49 00:04:15,300 --> 00:04:25,110 that you have a big list of companies there, and you've got this code here, or you can even use the search 50 00:04:25,110 --> 00:04:26,830 function of Yahoo. 51 00:04:27,270 --> 00:04:30,070 So these symbols are also called tickers. 52 00:04:30,180 --> 00:04:40,680 So you can enter a ticker here, a company name, so Apple and you see the ticker for Apple is AAPL. 53 00:04:41,010 --> 00:04:48,420 So you put that in there and then you want to specify the data source. 54 00:04:48,420 --> 00:04:53,720 So let's try Yahoo. You should get probably the same data. 55 00:04:53,850 --> 00:04:56,190 So Google gathers data, Yahoo gathers 56 00:04:56,210 --> 00:04:56,890 data 57 00:04:56,890 --> 00:04:58,800 on its sites. 58 00:04:58,890 --> 00:05:00,870 So that's not a big issue. 59 00:05:00,910 --> 00:05:03,740 I haven't tried much the data from Google. 60 00:05:03,760 --> 00:05:09,730 I have worked with Yahoo and I know that the maximum resolution, the temporal resolution that Yahoo 61 00:05:09,730 --> 00:05:11,830 offers is one day. 62 00:05:11,830 --> 00:05:16,180 So you can get a row for each day of the stock market data. 63 00:05:16,570 --> 00:05:22,180 You can also get weekly data. Then if you want you can easily reassemble these data into weekly 64 00:05:22,180 --> 00:05:23,930 or monthly with Pandas. 65 00:05:24,100 --> 00:05:25,630 So Yahoo. 66 00:05:25,640 --> 00:05:32,610 And then you have the start time, and the end time of your data. 67 00:05:32,980 --> 00:05:36,210 So let's define this here. 68 00:05:36,310 --> 00:05:45,860 Let's say the start variable, and that would be datetime.datetime. Actually we need to import the datetime first. 69 00:05:45,890 --> 00:05:48,070 Import datetime. 70 00:05:48,730 --> 00:05:52,630 And then we access the datetime class from the datetime library. 71 00:05:52,630 --> 00:06:01,820 So let's get data from 2016 and let's say March 1st. 72 00:06:01,840 --> 00:06:05,970 So this is the start time, and than the end time would be. 73 00:06:06,430 --> 00:06:13,790 So asyou see I'm using the datetime module to specify the date time of my data so. 74 00:06:13,900 --> 00:06:20,240 And the reason for that, the reason I'm putting these date times here is this the start and the end parameters 75 00:06:20,770 --> 00:06:24,610 they expect from you to enter a datetime data type. 76 00:06:24,940 --> 00:06:26,530 So you cannot just enter a string 77 00:06:26,530 --> 00:06:31,680 there that says 2016 and a coma, or an underscore, or anything. 78 00:06:31,720 --> 00:06:34,980 So you need to be strict there with your format. 79 00:06:35,170 --> 00:06:39,320 So let's say March and let's see 10 here. 80 00:06:39,520 --> 00:06:50,200 So for data for 10 days. Then you need to pass this variables here to these parameters, so starts and ends. That's 81 00:06:50,200 --> 00:06:51,520 it. 82 00:06:51,550 --> 00:06:55,420 So don't confuse the names of the parameters with the names of the variables. 83 00:06:55,660 --> 00:06:57,040 They just come to be the same here. 84 00:06:57,080 --> 00:07:02,180 Now we created these. 85 00:07:02,350 --> 00:07:09,760 And what you want to do now is you want to save the product in a variable. 86 00:07:10,390 --> 00:07:19,810 So I'm choosing to have a df variable, and actually if you check the type of all this expression 87 00:07:22,620 --> 00:07:24,300 well you will first get datetime. 88 00:07:24,300 --> 00:07:28,670 it's not defined because I didn't execute this line where I had it datetime. 89 00:07:28,730 --> 00:07:31,130 So execute that, now execute that again. 90 00:07:31,350 --> 00:07:36,380 And now you see that this object is actually a data frame. 91 00:07:36,390 --> 00:07:45,070 So the pandas_datareader, actually the datareader method of the data class of the pandas_datareader 92 00:07:45,070 --> 00:07:49,540 library produces a data frame, so a pandas data frame. 93 00:07:49,650 --> 00:07:59,400 So that means we can store that data frame in a variable. Remove the brackets and just print out the 94 00:07:59,400 --> 00:08:00,860 data frame here. 95 00:08:01,770 --> 00:08:02,840 And here we go. 96 00:08:02,850 --> 00:08:06,140 Let me close this window. 97 00:08:06,330 --> 00:08:09,470 And here's the data. Now, 98 00:08:09,470 --> 00:08:14,870 don't worry for now if you do not understand these attributes. I'll explain them in the next lecture 99 00:08:15,470 --> 00:08:20,820 because in this lecture I just wanted to know how we get around getting data with Python. 100 00:08:21,260 --> 00:08:24,940 So that is a product you get and you can export that in a csv file. 101 00:08:24,960 --> 00:08:26,690 So that's not a problem. 102 00:08:26,810 --> 00:08:35,240 And if you want to you can get data for a longer longer timeframe so you get a long data frame there. 103 00:08:35,690 --> 00:08:43,100 So from January to March. 1st of January and 10th of March. Of course you can change the company very 104 00:08:43,100 --> 00:08:47,530 easily to anything. Let me just put a random name there. 105 00:08:47,740 --> 00:08:52,230 There are quite a lot of companies out there so this should work. 106 00:08:52,540 --> 00:08:53,650 XRA. 107 00:08:54,140 --> 00:08:57,950 I have no idea what companies is this, but you get the idea. 108 00:08:57,950 --> 00:09:03,320 So it doesn't look as big as Apple, or even on Google. 109 00:09:03,320 --> 00:09:05,870 So you pass GOOG there. 110 00:09:06,420 --> 00:09:13,140 And this is a data for Google, and you can even get that from Google directly. 111 00:09:14,420 --> 00:09:26,450 So forth of January you have 743 for open. Let's see what Yahoo says about Google. 112 00:09:26,450 --> 00:09:27,350 So it says the same thing. 113 00:09:27,350 --> 00:09:32,410 That is about getting stock market data. 114 00:09:32,560 --> 00:09:38,650 Now I can show you how to get other kinds of data as well from the pandas_datareader library. 115 00:09:39,020 --> 00:09:41,680 And I would prefer to point you to the documentation. 116 00:09:41,750 --> 00:09:45,740 I don't want to go through every dataset there. 117 00:09:45,730 --> 00:09:51,740 This is a very clear documentation so I'm sure you'll have no problem going through this. 118 00:09:51,770 --> 00:09:59,330 So here you have the data from Yahoo which we just consumed. 119 00:09:59,900 --> 00:10:02,300 Then you can have data from Google. 120 00:10:02,300 --> 00:10:10,370 So you pass Google there just as I did here. I passed the name of a parameter explicitly there, but you can also 121 00:10:10,730 --> 00:10:13,340 pass them in the correct order like this. 122 00:10:13,340 --> 00:10:19,130 So the company name, the datasource name, start and end, and you'll be fine. 123 00:10:19,130 --> 00:10:21,500 So it is the same thing. 124 00:10:21,680 --> 00:10:25,650 Then you have data from FRED, and Fama/French. 125 00:10:26,000 --> 00:10:30,170 And interestingly you have data from World Bank as well. 126 00:10:30,230 --> 00:10:38,200 So instead of doing from pandas_datareader import data as we did here, you do from pandas_datareader 127 00:10:38,220 --> 00:10:39,160 import 128 00:10:39,380 --> 00:10:48,610 WB so World Bank, and then you search for data sets of GDP per capita and and also other attributes there. 129 00:10:49,390 --> 00:10:55,040 And you should use this download function and that gets you data in a data frame as well. 130 00:10:56,710 --> 00:11:00,610 So that's about World Bank and you have other data sets as well. 131 00:11:00,630 --> 00:11:09,070 So you have OECD statistics there, Eurostats and Edgard index. If you ever need them you know 132 00:11:09,070 --> 00:11:15,270 that they are here in this URL, so in the pandas_datareader documentation. 133 00:11:15,820 --> 00:11:19,600 So that's what I wanted to teach you in this lecture. 134 00:11:19,760 --> 00:11:22,580 I will move on by focusing on the stock market data. 135 00:11:22,590 --> 00:11:27,180 So I'll go quickly through this data explaining them what they mean. 136 00:11:27,190 --> 00:11:30,920 So we've got quite a lot of interesting things to do. 137 00:11:30,970 --> 00:11:31,980 Let's move on then.