If you search for data analysis workflows for research there are lots of blog posts on using R + databases + git, etc. While in some cases I may end up working with a combination like that, it’s much more likely that reality is closer to a bunch of emailed Excel or CSV files.

Some may argue that one should move the whole group of collaborators to work the right way. In practice, well, not everyone has the interest and/or the time to do so. In one of our collaborations we are dealing with a trial established in 2009 and I was tracking a field coding mistake (as in happening outdoors, doing field work, assigning codes to trees), so I had to backtrack where the errors were introduced. After checking emails from three collaborators, I think I put together the story and found the correct code values in a couple of files going back two years.

The new analysis lives in an RStudio project with the following characteristics:

Folder in Dropbox, so it’s copied in several locations and it’s easy to share.

Excel or CSV files with their original names (warts and all), errors, etc. Resist the temptation to rename the files to sane names, so it’s easier to track back the history of the project.

R code

Important part: text file (Markdown if you want) documenting the names of the data files, who & when they sent it to me.

I was having a conversation about dropping the minimum threshold (currently 5% of the vote) for political parties to get representation in Parliament. The obvious question is how would seat allocation change, which of course involved a calculation. There is a calculator in the Electoral Commission website, but trying to understand how things work (and therefore coding) is my thing, and the Electoral Commission has a handy explanation of the Sainte-Laguë allocation formula used in New Zealand. So I had to write my own seat allocation function:

# National Labour NZ First Green ACT Opportunities Māori

# [1,] 1 2 6 9 98 22 46

# [2,] 3 4 19 26 NA 67 NA

# [3,] 5 7 33 42 NA 112 NA

# [4,] 8 11 47 59 NA NA NA

# [5,] 10 13 60 77 NA NA NA

# [6,] 12 15 73 93 NA NA NA

# [7,] 14 17 86 110 NA NA NA

# [8,] 16 21 100 NA NA NA NA

# [9,] 18 24 113 NA NA NA NA

#[10,] 20 27 NA NA NA NA NA

# ...

In our current setup The Opportunities and Māori parties did not reach the minimum threshold (nor won an electorate as ACT violating the spirit of the system), so did not get any seats. Those 4 seats that would have gone to minor parties under no threshold ended up going to National and Labour (2 each). It sucks.

I was reading an article about the results of our latest elections where I was having a look at the spatial pattern for votes in my city.

I was wondering how would I go over obtaining the data for something like that and went to the Electoral Commission, which has this neat page with links to CSV files with results at the voting place level. The CSV files have results for each of the candidates in the first few rows (which I didn’t care about) and at the party level later in the file.

As I could see it I needed to:

Read the Electoral Commission website and extract the table that contains the links to all CSV files.

Read each of the files and i- extract the electorate name, ii- skipping all the candidates votes, followed by iii-reading the party vote.

Remove sub-totals and other junk from the files.

Geocode the addresses

Use the data for whatever else I wanted (exam question anyone?).

So I first loaded the needed packages and read the list of CSV files:

1

2

3

4

5

6

7

8

9

10

11

12

13

14

library(magrittr)

library(tidyverse)

library(rvest)

library(stringr)

library(ggmap)

# Extract list of CSV file names containing voting place data

Then wrote a couple of functions to, first, read the whole file, get the electorate name and, second, detect where the party vote starts to keep from that line onwards. Rather than explicitly looping over the list of CSV file names, I used map_dfr from the purrr package to extract the data and join all the results by row.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

get_electorate<-function(row){

row%>%str_split(pattern=',')%>%

unlist()%>%.[1]%>%str_split(pattern='-')%>%

unlist()%>%.[1]%>%str_trim()->elect

return(elect)

}

# Function to read only party-level votes from voting places

Cleaning the data and summarising by voting place (as one can vote for several electorates in a single place) is fairly straightforward. I appended the string Mobile to mobile teams that visited places like retirement homes, hospitals, prisons, etc:

Geolocation is the not-working-very-well part right now. First, I had problems with Google (beyond the 1,000 places limit for the query). Then I went for using the Data Science Kit as the source but, even excluding the mobile places, it was a bit hit and miss for geolocation, particularly as the format of some address (like corner of X and Y) is not the best for a search.

In addition, either of the two sources for geolocation work really slowly and may produce a lot of output. Using sink() could be a good idea to not end up with output for roughly 3,000 queries. I did try the mutate_geocode() function, but didn’t work out properly.

I was reading a piece by Graeme Edgeler who, near the end, asked “Where are New Zealand’s bellwether electorates?”. I didn’t know where the data came from or how was the “index of disproportionality for each electorate” calculated, but I saw it mostly as an opportunity to whip up some quick code to practice the use of R and look at other packages that play well with the tidyverse.

The task can be described as: fetch Wikipedia page with results of the 2014 parliamentary election, extract the table with results by electorate, calculate some form of deviation from the national results, get the top X electorates with lowest deviation from national results.

A web search revealed that this page contains a whole bunch of results for the 2014 election and that the specific results I’m interested in are in table number 17 of the list created by html_nodes('table'). Besides the tidyverse, I needed the packages rvest for web scraping, magrittr for using %<>% (pipe and assign to original data frame) and lucid for pretty printing the final table.

Rather than reading the national results directly from Wikipedia I just typed them in code, as I already had them from some other stuff I was working on. My measure of “disproportionality for each electorate” was as sophisticated as the sum of squared deviations.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

# Results for National, Labour, Green, NZ First, Conservative, Internet Mana & Māori

I’m sure there must be a ‘more idiomatic’ way of doing the squared deviation using the tidyverse. At the same time, using apply came naturally in my head when writing the code, so I opted for keeping it and not interrupting the coding flow. The results are pretty similar to the ones presented by Graeme in his piece.

I’m getting increasingly comfortable with this mestizo approach of using the tidyverse and base R for completing tasks. Whatever it takes to express what I need to achieve quickly and more or less in a readable way.

Often I need to apply a function by groups of observations; sometimes, that function returns more than a single number. It could be something like for each group fit a distribution and return the distribution parameters. Or, simpler for the purposes of this exploration, calculate and return a bunch of numbers.

1

2

3

4

5

6

7

describe_c<-function(x){

mn<-mean(x,na.rm=TRUE)

dev<-sd(x,na.rm=TRUE)

n<-sum(!is.na(x))

cv<-dev/mn*100

return(c(mean=mn,sdev=dev,count=n,coefvar=cv))

}

If I have a data frame called field_data, with family codes (trees with the same parents, codes have been changed to protect the innocent) and stem diameters (in mm), I could do the following in base R:

# To continue processing it might be better to convert

# to a data frame

summary_one<-data.frame(summary_one)

And if I need to do this for several variables, I will need to merge each of these matrices in a data frame.

Continuing with my experimentation with the tidyverse, I was wondering how to get the above going with dplyr et al. After failing a few times I asked the question in Twitter and got a number of helpful replies.

One of the keys is that dplyr can store a list result from a function. Modifying my toy function is pretty straightforward, and now looks like:

And we can check the contents of summary_two to see we have a list in which each element contains 4 values:

1

2

3

4

5

6

7

8

9

10

head(summary_two)

# A tibble: 6 x 2

# family model

# <fctr> <list>

# 1 A <dbl [4]>

# 2 B <dbl [4]>

# 3 C <dbl [4]>

# 4 D <dbl [4]>

# 5 E <dbl [4]>

# 6 F <dbl [4]>

We still need to extract the elements of each element of the list and assign them to a variable name. Using map from the purrr package is pretty straightforward in this case, and we can extract the values either using their names or their position in the element.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

summary_two%>%

mutate(mn=map_dbl(model,'mean'),

sd=map_dbl(model,'sdev'),

n=map_dbl(model,'count'),

cv=map_dbl(model,4))%>%head

# A tibble: 6 x 6

# family model mn sd n cv

# <fctr> <list> <dbl> <dbl> <dbl> <dbl>

# 1 A <dbl [4]> 190.8306 23.71290 425 12.42615

# 2 B <dbl [4]> 190.1111 25.46554 396 13.39508

# 3 C <dbl [4]> 188.2646 27.39215 461 14.54981

# 4 D <dbl [4]> 189.2668 25.16330 431 13.29514

# 5 E <dbl [4]> 183.5238 19.70182 21 10.73530

# 6 F <dbl [4]> 183.1250 28.82377 24 15.73994

I’m still playing with ideas to be lazier at extraction time. An almost abhorrent idea is to provide the output as character for posterior type conversion, as in:

One of the main uses for R is for exploration and learning. Let’s say that I wanted to learn simple linear regression (the bread and butter of statistics) and see how the formulas work. I could simulate a simple example and fit the regression with R:

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

library(arm)# For display()

# Simulate 5 observations

set.seed(50)

x<-1:5

y<-2+3*x+rnorm(5,mean=0,sd=3)

# Fit regression

reg<-lm(y~x,dat)

display(reg)

# lm(formula = y ~ x, data = dat)

# coef.est coef.se

# (Intercept) 3.99 3.05

# x 2.04 0.92

# ---

# n = 5, k = 2

# residual sd = 2.91, R-Squared = 0.62

# Plot it

plot(y~x)

abline(coef(reg))

The formulas for the intercept (\(b_0\)) and the slope (\(b_1\)) are pretty simple, and I have been told that there is a generic expression that instead uses matrices.

\(b_1 = \frac{\sum{x y} – n \bar{x} \bar{y}}{\sum{x x} – n \bar{x}^2}\)
\(b_0 = \bar{y} – b_1 \bar{x}\)

Funnily enough, looking at the matrices we can see similar sums of squares and crossproducts as in the formulas.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

X<-model.matrix(reg)

# (Intercept) x

# 1 1 1

# 2 1 2

# 3 1 3

# 4 1 4

# 5 1 5

# attr(,"assign")

# [1] 0 1

t(X)%*%X

# (Intercept) x

# (Intercept) 5 15

# x 15 55

# So X`X contains bits and pieces of the previous formulas

length(x)

# [1] 5

sum(x)

# [1] 15

sum(x*x)

# [1] 55

# And so does X`y

t(X)%*%y

# [,1]

# (Intercept) 50.61283

# x 172.27210

sum(y)

# [1] 50.61283

sum(x*y)

# [1] 172.2721

# So if we combine the whole lot and remember that

# solves calculates the inverse

solve(t(X)%*%X)%*%t(X)%*%y

# [,1]

# (Intercept) 3.992481

# x 2.043362

But I have been told that R (as most statistical software) doesn’t use the inverse of the matrix for estimating the coefficients. So how does it work?

If I type lm R will print the code of the lm() function. A quick look will reveal that there is a lot of code reading the arguments and checking that everything is OK before proceeding. However, the function then calls something else: lm.fit(). With some trepidation I type lm.fit, which again performs more checks and then calls something with a different notation:

1

z<-.Call(C_Cdqrls,x,y,tol,FALSE)

This denotes a call to a C language function, which after some searching in Google we find in a readable form in the lm.c file. Another quick look brings more checking and a call to Fortran code:

1

2

3

F77_CALL(dqrls)(REAL(qr),&n,&p,REAL(y),&ny,&rtol,

REAL(coefficients),REAL(residuals),REAL(effects),

&rank,INTEGER(pivot),REAL(qraux),work);

which is a highly tuned routine for QR decomposition in a linear algebra library. By now we know that the general matrix expression produces the same as our initial formula, and that the R lm() function does not use a matrix inverse but QR decomposition to solve the system of equations.

One of the beauties of R is that brought the power of statistical computing to the masses, by not only letting you fit models but also having a peek at how things are implemented. As a user, I don’t need to know that there is a chain of function calls initiated by my bread-and-butter linear regression. But it is comforting to the nerdy me, that I can have a quick look at that.

All this for free, which sounds like a very good deal to me.

I started using R ages ago and have happily lived in mostly-base-R for data manipulation. Once in a while I move to something that makes a big difference, like ggplot2 in 2010 or Rmarkdown in 2015, but the set of packages I use for data + plotting hasn’t seen many changes. I have to confess that, meanwhile, I have tested quite a few approaches on the analytics side of things (last year was the turn of Bayesian for me).

Last week, I decided to learn more about the tidyverse, thinking of using it more with forestry postgrad students. Now, there is no lack of tutorials, reviews, documentation, etc. for the tidyverse, but most writing shows a final version of the code, without exposing the thinking and dead ends that go behind it. In this post I show how my code was changing, both after reading a few pieces of documentation and, mostly, from feedback I got from Hadley Wickham and Michael MacAskill via this Kiwi Twitter thread. This post shows minor differences in variable names from that thread, as I changed a few things while reading the files.

The problem is as follows: I have two data frames with trial assessments. Frame one, called early, covers trees at ages 5, 7 and 8 years (although ages are in months rather than years). Frame two, called late, covers trees at age 20 years. Yes, it takes a while working with trees.

We want to keep only age 8 years (96 months) from early and want to split a code into two variables, as well as convert a variable from numeric to character. In late we want to create a tree family code, based on a set of rules to connect field codes to the pedigree of trees. Did I mention that I work breeding trees?

Finally, we want to merge all the assessments from age 8 with the assessment at age 20 for the same trees.

Rather than showing the final version of the code, it is much more interesting to show its evolution, also including how I would have done this in base R. I’m omitting the reading of the file and boring case conversion of variable names, etc.

In base R, I would probably do something like this (I’m using the stringr package just to make my life easier):

I used multiple instances of mutate because I thought it would be easier to read. The use of map instead of sapply is cool, particularly when one starts looking at more advanced features. Comments from the crowd in Twitter: mutate the whole lot in a single statement (although Hadley pointed out that there was no performance penalty by using them separately) and try using case_when to make nested ifelse easier to understand. The comments on map went in two directions: either use map_chr as a clearer and safer alternative, or just use dplyr‘s separate function. The first option for early would look like:

However, I ended up going with the final version that used separate, which is easier on the eye and faster, for a final version that looks like this:

1

2

3

4

5

early%>%

filter(age==96)%>%

separate(section,c('rep','sets'),sep='_')%>%

mutate(tree=as.character(`tree position`),

genotype=as.character(1:nrow(.)+10^6))->early_8

So we filter early, separate the single set code into two variables (rep and sets) and create a couple of variables using mutate (one is a simple type conversion to character, while the other is a code starting at 1,000,000).

Some (many, most?) people may disagree with my use of right assign, which I love. Surely one could use either left assign or %<>% from the maggrittr package. By the way, why do I have to explicitely load magrittr (instead of relying on tidyverse) to access %<>%?

And this is how I go about learning new things: lots of false starts, often working with small examples (I used a few to check how left_join was working), lots of searching for explanations/tutorials (thanks to everyone who has written them) and asking in Twitter. If you are just starting programming, in any language, do not feel intimidated by cool looking code; most of the time it took many iterations to get it looking like that.

I was attending a course of Bayesian Statistics where this problem showed up:

There is a number of individuals, say 12, who take a pass/fail test 15 times. For each individual we have recorded the number of passes, which can go from 0 to 15. Because of confidentiality issues, we are presented with rounded-to-the-closest-multiple-of-3 data (\(\mathbf{R}\)). We are interested on estimating \(\theta\) of the Binomial distribution behind the data.

Rounding is probabilistic, with probability 2/3 if you are one count away from a multiple of 3 and probability 1/3 if the count is you are two counts away. Multiples of 3 are not rounded.

We can use Gibbs sampling to alternate between sampling the posterior for the unrounded \(\mathbf{Y}\) and \(\theta\). In the case of \(\mathbf{Y}\) I used:

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

# Possible values that were rounded to R

possible<-function(rounded){

if(rounded==0){

options<-c(0,1,2)

}else{

options<-c(rounded-2,rounded-1,rounded,

rounded+1,rounded+2)

}

return(options)

}

# Probability mass function of numbers rounding to R

# given theta

prior_y<-function(options,theta){

p<-dbinom(options,15,prob=theta)

return(p)

}

# Likelihood of rounding

like_round3<-function(options){

if(length(options)==3){

like<-c(1,2/3,1/3)}

else{

like<-c(1/3,2/3,1,2/3,1/3)

}

return(like)

}

# Estimating posterior mass function and drawing a

# random value of it

posterior_sample_y<-function(R,theta){

po<-possible(R)

pr<-prior_y(po,theta)

li<-like_round3(po)

post<-li*pr/sum(li*pr)

samp<-sample(po,1,prob=post)

return(samp)

}

While for \(theta\) we are assuming a vague \(mbox{Beta}(alpha, eta)\), with \(alpha\) and \(eta\) equal to 1, as prior density function for \(theta\), so the posterior density is a \(mbox{Beta}(alpha + sum Y_i, eta + 12*15 – sum Y_i)\).

1

2

3

4

5

## Function to sample from the posterior Pr(theta | Y, R)

I’ve been having a conversation for a while with @kamal_hothi and @aschiff on maps, schools, census, making NZ data available, etc. This post documents some basic steps I used for creating a map on ethnic diversity in schools at the census-area-unit level. This “el quicko” version requires 3 ingredients:

Census area units shape files (available from Statistics New Zealand for free here).

P.S. Using the diversity-index-census-area-unit.csv and joining it with the shapefile in QGIS one can get something prettier (I have to work on matching the color scales):

Map rendering is so much faster in QGIS than in R! Clearly the system has been optimized for this user case.

When I write code I’m often amazed by the direction of the statements. I mean, we read and write left-to-right except when we assign statements to a variable. So here we are, doing our business, slowly working in a sentence and, puff!, we get this insight from the future and we assign it to our past, on the left. I commented this on Twitter and @fadesingh pointed me to this artistic creation of right-to-left programming language, except that the assign was left-to-right.

R has a right-assign operator (->) and I was thinking, how would it look if I wrote all the assigns to the left, using an adaptation of my previous post.