TDM 10100: R Project 13 — 2024

First load: library(data.table) so that you have the fread function available, and also options(repr.matrix.max.cols=50) so that you can see 50 columns.

Also load: options(jupyter.rich_display = T) so that you can draw maps in Jupyter Lab.

For this project, we need two maps packages: library(leaflet) and library(sf)

For this first question, we simply make a simple map, with three latitude and longitude values.

We call our data frame testDF, and it is important that we name the columns:

The whole data frame looks like this, just three rows and two columns:

Now we can define the points that we want to plot on a map:

and we can render the map. We make each dot have radius 1, but you are welcome to change the radius if you want to:

Now load the Craiglist data as follows:

Examine the head of myDF and also the dim of myDF, and see which columns are the state and the long and the lat columns.

Now read ALL of the rows of the data set into a new data frame, but only the three columns called state and long and lat (the other columns will not be needed).

Make a subset of this new data frame, satisfying 3 conditions, namely, the state variable indicates that the data is from Indiana, and the long and lat values are not missing.

You should now have a data frame with 3 columns and 5634 rows.

Now make a plot of the data frame that you created in question 2, using these two lines of R:

Please note that, with Craigslist, people can list items from anywhere in the country. So there are some items outside Indiana, even though we selected only the items that are supposed to be from the State of Indiana. BUT fortunately, you will see that most people’s listings are accurate. In other words, if you zoom in and out on the map, you will see that most of the dots appear in Indiana.

In question 4 and question 5, we will verify the path of the Thanksgiving parade in New York City from 2015, as shown on this image:

www.bizjournals.com/newyork/news/2015/11/25/thanksgiving-is-tomorrow-but-parade-re-routes-and.html

You can import the New York taxi cab data from November 2015 as follows:

(The tz="" indicates that the time zone is not given in this data.)

Make a new data frame called thanksgivingdayDF by using the subset function, with the option grepl("2015-11-26", tpep_pickup_datetime) to extract the rows of the data from Thanksgiving day. Your new data frame should have 242393 rows and 19 columns.

Create two times in R, one for the start of the parade, and one for the end of the parade:

Now make a vector of times (converting the pickup times from the taxi cab rides from strings into times).

Finally, make a new data frame called finalDF from the data frame thanksgivingdayDF, using the subset function with the condition (mytimes > paradestart) & (mytimes < paradeend).

Your data frame finalDF should have 28704 rows.

If you examine the head of finalDF, you see that the latitude values are called pickup_latitude and pickup_longitude.

We want them to be called lat and long instead, so we can make a new data frame as follows:

Finally, plot the latitude and longitude values from testDF using a smaller radius than you used in Question 1 and Question 3. We suggest radius=.1.

You will notice that taxi cabs were unable to pickup passengers on the route of the Thanksgiving Day parade because those roads were closed. Please zoom into the map and verify this, comparing your map to the parade route map:

www.bizjournals.com/newyork/news/2015/11/25/thanksgiving-is-tomorrow-but-parade-re-routes-and.html