Fells Stats Know Your Data


New Book in Text Analysis for R

Focus for books on R tend to be highly focused on either statisticians or programmers. There is a dearth of material to assist those in typically less quantitative field access the powerful tools in the R ecosystem. Enter Text Analysis with R for Students of Literature.

I haven't done a deep read of the book, but the goal of opening up R to the Literature community is a laudable one. Plus the wordcloud package gets a shout out!

Filed under: R, wordcloud No Comments

New Version of the OpenStreetMap R Pacakge

A new version of the OpenStreetMap package has been released to CRAN. OpenStreetMap 0.3.3 contains several minor improvements. I've removed the CloudMade tile set types, as they seem to have gone out of business. MapQuest has also been removed as they have moved to a new API. The mapbox type has been updated to use their new API.

The most important update is the ability to use custom tile servers. Suppose you have a server using the tile map service specification with access urls looking something like


where {z} is the map's integer zoom level, {x} is the tile's integer x position and {y} is the tile's integer y position. You may pass this url into openmap's type argument to obtain a map using this tileset.

For example, if you need to access MapQuest's tilesets, which now require a developer API key, you can use the new custom tileserver facility to access them. Below is an example with my (free) API key.


ul <- c(40.9,-74.5)
lr <- c(40.1,-73.2)
url <- "https://a.tiles.mapbox.com/v4/mapquest.streets-mb/{z}/{x}/{y}.png?access_token=pk.eyJ1IjoibWFwcXVlc3QiLCJhIjoiY2Q2N2RlMmNhY2NiZTRkMzlmZjJmZDk0NWU0ZGJlNTMifQ.mPRiEubbajc6a5y9ISgydg"
map <- openmap(ul,lr,minNumTiles=4, type=url)

url <- "https://a.tiles.mapbox.com/v4/mapquest.satellitenolabels/{z}/{x}/{y}.png?access_token=pk.eyJ1IjoibWFwcXVlc3QiLCJhIjoiY2Q2N2RlMmNhY2NiZTRkMzlmZjJmZDk0NWU0ZGJlNTMifQ.mPRiEubbajc6a5y9ISgydg"
map <- openmap(ul,lr,minNumTiles=4, type=url)

url <- "https://a.tiles.mapbox.com/v4/mapquest.satellite/{z}/{x}/{y}.png?access_token=pk.eyJ1IjoibWFwcXVlc3QiLCJhIjoiY2Q2N2RlMmNhY2NiZTRkMzlmZjJmZDk0NWU0ZGJlNTMifQ.mPRiEubbajc6a5y9ISgydg"
map <- openmap(ul,lr,minNumTiles=4, type=url)

url <- "https://a.tiles.mapbox.com/v4/mapquest.dark/{z}/{x}/{y}.png?access_token=pk.eyJ1IjoibWFwcXVlc3QiLCJhIjoiY2Q2N2RlMmNhY2NiZTRkMzlmZjJmZDk0NWU0ZGJlNTMifQ.mPRiEubbajc6a5y9ISgydg"
map <- openmap(ul,lr,minNumTiles=4, type=url)

url <- "https://a.tiles.mapbox.com/v4/mapquest.light/{z}/{x}/{y}.png?access_token=pk.eyJ1IjoibWFwcXVlc3QiLCJhIjoiY2Q2N2RlMmNhY2NiZTRkMzlmZjJmZDk0NWU0ZGJlNTMifQ.mPRiEubbajc6a5y9ISgydg"
map <- openmap(ul,lr,minNumTiles=4, type=url)








Insert and Remove Performance of Boost’s flat_set v.s. std::set

The standard way to represent an ordered set of numbers is with a binary tree. This offers a good mix of performance properties as the number of elements gets large. In particular it offers O(log(n)) operations insertion/deletion, O(log(n)) operations to find an element. Finding the ith element of the set takes more time, O(n) operations for random access compared to a vector, which has O(1) complexity for that task.

In C++ std::set implements an ordered set as a binary tree, hiding all the ugly details for you. But std::set is not always the right structure to use. Unless you are doing lots of inserts and removals in large sets, binary trees can be inefficient because they have to do things like pointer referencing, and the data is not stored in a continuous block leading to cache misses. This is why some have declared that you shouldn't use set.

An alternative is to use a sorted std::vector as the set data structure. This improves the performance of random access to O(1), and worsens the performance of insertion/deletion to O(n). There is a handy drop in replacement for std::set in boost called flat_set. You can mostly take any code using set and switch it to flat_set with no changes to the logic.

Recently I was in a situation where I had performance critical code using many small sets (typically between 0 and 20 elements). This code does lots of insertions and deletions, so one would initially think that flat_set is not a good option with its O(n) complexity, but remember that complexity is an asymptotic statement as n grows, and I was relatively certain that my n was small. So which should be used? The only way to find out was to do some simulations.

For each n I generated a set<int> and flat_set<int> with equally spaced integers from 0 to 100,000,000, then I inserted and removed 500,000 random integers in the same range and recorded the timings. The compiler optimization was set to high ( -O3 ).

size  flat_set  std::set
2     0.02791   0.070053
4     0.031647  0.07919
8     0.038431  0.08474
16    0.0528    0.091744
32    0.0697    0.104424
64    0.085957  0.129225
128   0.1176    0.129537
256   0.15825   0.138755
512   0.253153  0.148642
1024  0.401831  0.156223
2048  0.718302  0.166177
4096  1.35207   0.176593
8192  2.5926    0.19331

Times are in seconds here, so for small sets (2-16 elements) flat_set is twice as fast, and beats std::set all the way up though 128 elements. By 4096 elements we are paying the asymptotic cost however and flat_set is >10x slower. flat_set is vector backed, so we know that it will be much faster at other operations like random access, iteration and find because it is in a contiguous block of memory. The surprising thing is that it is even faster at insertions and deletions provided the set size is modest.

If you are an R user, you can use flat_set very easily now with the new BH package. Simply add it as a linkingTo in your package and Bob is your uncle. Below is the code that I used for the simulations (it uses Rcpp but that can easily be taken out).

        #include <boost/container/flat_set.hpp>
        #include <set>
        #include <ctime>
        #include <Rcpp.h>
        std::clock_t start;
	double d1,d2;
	boost::container::flat_set<int> fs;
	std::set<int> ss;
	int n = 500000;
	int max = 100000000;
	for(int i = 1;i<14;i++){
		int s = round(pow(2.0,i));
		for(int j=0;j<s;j++){
			int val = round(((j+1)/(double)s)*max);//floor(Rf_runif(0.0,max));
		start = std::clock();
		for(int j=0;j<n;j++){
			int rand = floor(Rf_runif(0.0,max));
			bool ins = fs.insert(rand).second;
		d1 += ( std::clock() - start ) / (double) CLOCKS_PER_SEC;

		start = std::clock();
		for(int j=0;j<n;j++){
			int rand = floor(Rf_runif(0.0,max));
			bool ins = ss.insert(rand).second;
		d2 += ( std::clock() - start ) / (double) CLOCKS_PER_SEC;
		std::cout << s << ", " << d1 << ", " << d2 << "," << std::endl;



Filed under: C++, R, Rcpp No Comments

The OpenStreetMap Package Opens Up

A new version of the OpenStreetMap package is now up on CRAN, and should propagate to all the mirrors in the next few days. The primary purpose of the package is to provide high resolution map/satellite imagery for use in your R plots. The package supports base graphics and ggplot2, as well as transformations between spatial coordinate systems.

The bigest change in the new version is the addition of dozens of tile servers, giving the user the option of many different map looks, including those from Bing, MapQuest and Apple.

nm = c("osm", "maptoolkit-topo",
"waze", "mapquest", "mapquest-aerial",
"bing", "stamen-toner", "stamen-terrain",
"stamen-watercolor", "osm-german", "osm-wanderreitkarte",
"mapbox", "esri", "esri-topo",
"nps", "apple-iphoto", "skobbler",
 "opencyclemap", "osm-transport",
"osm-public-transport", "osm-bbike", "osm-bbike-german")

for(i in 1:length(nm)){
	map = openmap(c(lat= 38.05025395161289,   lon= -123.03314208984375),
			c(lat= 36.36822190085111,   lon= -120.69580078125),

You can also render maps from cloudmade.com which hosts tons of map tilings. Simply set the "type" parameter to "cloudmade-<id>" where <id> is the cloudmade identifier for the map you want to use. Here is a sample:

nm = c("cloudmade-2","cloudmade-999","cloudmade-998",
#setCloudMadeKey("<your key>")
for(i in 1:length(nm)){
	map = openmap(c(lat= 38.05025395161289,   lon= -123.03314208984375),
			c(lat= 36.36822190085111,   lon= -120.69580078125),

Maps are initially put in a sperical mercator projection which is the standard for most (all?) map tiling systems, but can easily be translated to long-lat (or any other projection) using the openproj function. Maps can be plotted in ggplot2 using the autoplot function.

mapLatLon = openproj(map)

The package also has a Java GUI to help with map type selection, and specification of coordinates to bound your map. clicking on the map will give you the latitude and longitude of the point clicked.


Probably the main alternative to OpenStreetMap is the ggmap package. ggmap is an excellent package, and it is somewhat unfortunate that there is a significant duplication of effort between it and OpenStreetMap. That said, there are some differences that may help you decide which to use:

Reasons to favor OpenStreetMap:

  • More maps: OpenStreetMap supports more map types.
  • Better image resolution: ggmap only fetches one png from the server, and thus is limited to the resolution of that png, whereas OpenStreetMap can download many map tiles and stich them together to get an arbitrarily high image resolution.
  • Transformations: OpenStreetMap can be used with any map coordinate system, whereas ggmap is limited to long-lat.
  • Base graphics: Both packages support ggplot2, but OpenStreetMap also supports base graphics.
Reasons to favor ggmap:
  • No Java dependency: ggmap does not require Java to be installed.
  • Geocoding: ggmap has functions to do reverse geo coding.
  • Google maps: While OpenStreetMap has more map types, it currently does not support google maps.





Interview by DecisionStats

Ajay Ohri interviewed
on his popular DecisionStats blog. Topics discussed
ranged widely from Fellows
, to Deducer, to statnet, to Poker A.I.,
to Big Data.    

Filed under: Deducer, R No Comments

Quickly Profiling Compiled Code within R on the Mac

This is a quick note on profiling your compiled code on the mac. It is important not to guess when figuring out where the bottlenecks in your code are, and for this reason, the R manual has several suggestions on how to profile compiled code running within R. All of the methods are platform dependent, with linux requiring command line tools (sprof, and oprofile). On the mac there are some convenient GUI tools built in to help you out.

Our test code is drawn from the ergm package, and fits a network model to a small graph. This model is fit using markov chain monte carlo, so we expect most of the time to be spent drawing MCMC samples.

flomarriage = network(flo,directed=FALSE)
flomarriage %v% "wealth" = c(10,36,27,146,55,44,20,8,42,103,48,49,10,48,32,3)
for(i in 1:10) gest = ergm(flomarriage ~ kstar(1:2) + absdiff("wealth") + triangle)

The quickest way to get an idea of where your program is spending time is to open Activity Monitor.app and select the R process while the above code is running. Then click sample, which should result in something like:

In order to get down to the interesting code, you'll have to wade though a bunch of recursive Rf_eval and do_begin calls used by R. Finally though we see the MCMCSample function in ergm.so which contains almost all of the samples. So it is as we expected, the MCMC sampling takes up all the time. Underneeth that we see calls to:

  • ChangeStats: calculates the network statistics
  • MH_TNT: calculates the metropolis proposal
  • ToggleEdge: alters the network to add (or remove) an edge

But how much these contribute is a bit difficult to see here. Enter the Time Profiler in Instruments.app. You can launch this program from within Xcode.app under Xcode - Open Developer Tool - Instruments. Once Instruments is open, click Library and drag the Time Profiler into the program. Then, while the R code is running, select 'Attach to Process' - R and click record. You should get something like:
Like in the Activity Monitor, we have to drill down past all of the recursive Rf_eval calls, but once we do, we see that ChangeStats is taking 44.6% of the time, MH_TNT takes 30.9% and ToggleEdges takes 11.7%. We could further expand these out to identify possible targets for optimization within each of these functions.

So there you have it. Apple makes it really easy to profile your compiled code. One issue that I ran into is to make sure that your Xcode is up-to-date and that Instruments is version 4.0 or above. I experienced some crashes profiling R with older versions.

Filed under: R No Comments

Great Infographic

This is a really great exposition on an
infographic. Note that the design elements and "chart junk" serve
to better connect and communicate the data to the viewer. The
choice not to go with pie charts for the first set of plots is a
good one. The drawbacks of polar representations of proportions is
very well know to statisticians, but designers often feel that
circular elements are more pleasing in their infographics. This
bias toward circular representations is sometimes justifiable.
After all, if your graphic is pretty then no one will take the time
to look at it.
Take a look:


Filed under: Uncategorized No Comments

Climate: Misspecified

I'm usually quite a big fan of the content syndicated on R-Bloggers (as this post is), but I came across a post yesterday that was as statistically misguided as it was provocative. In this post, entitled "The Surprisingly Weak Case for Global Warming," the author (Matt Asher) claims that the trend toward hotter average global temperatures over the last 130 years is not distinguishable from statistical noise. He goes on to conclude that "there is no reason to doubt our default explaination of GW2 (Global Warming) - that it is the result of random, undirected changes over time."

These are very provocative claims which are at odds with the vast majority of the extensive literature on the subject. So this extraordinary claim should have a pretty compelling analysis behind it, right?...

Unfortunately that is not the case. All of the author's conclusions are perfectly consistant with applying an unreasonable model, inappropriate to the data. This in turn leads him to rediscover regression to the mean. Note that I am not a climatologist (neither is he), so I have little relevant to say about global warming per se, rather this post will focus on how statistical methodologies should pay careful attention to whether the data generation process assumed is a reasonable one, and how model misspecification can lead to professional embarrassment.

His Analysis

First, let's review his methodology. He looked at the global temperature data available from NASA. It looks like this:

Average global temperatures (as deviations from the mean) with cubic regression

He then assumed that the year to year changes are independent, and simulated from that model, which yielded:

Here the blue lines are temperature difference records simulated from his model, and the red is the actual record. From this he concludes that the climate record is rather typical, and consistant with random noise.

A bit of a fly in the ointment though is that he found that his independence assumption does not hold. In fact he finds a negative correlation between one years temperature anomaly and the next:

Any statistician worth his salt (and indeed several of the commenters noted) that this looks quite similar to what you would see if there were an unaccounted for trend leading to a regression to the mean.

Bad Model -> Bad Result

The problem with using an autoregressive model here is that it is not just last year's temperatures which determine this year's temperatures. Rather, it would seem to me as a non-expert, that temperatures from one year are not the driving force for temperatures for the next year (as an autoregressive model assumes). Rather there are underlying planetary constants (albedo and such) that give a baseline for what the temperature should be, and there is some random variation which cause some years to be a bit hotter, and some cooler.

Remember that first plot, the one with the cubic regression line. Let's assume that data generation process is from that regression line, with the same variance of residuals. We can then simulate from the model to create an fictitious temperature record. The advantage of doing this is that we know the process that generated this data, and know that there exists a strong underlying trend over time.

Simulated data from a linear regression model with cubic terms

If we fit a cubic regression model to the data, which is the correct model for our simulated data generation process, it shows a highly significant trend.

              Sum Sq  Df F value    Pr(>F)
poly(year, 3)  91700   3  303.35 < 2.2e-16 ***
Residuals      12797 127

We know that this p-value (essentially 0) is correct because the model is the same as the one generating the data, but if we apply Mr. Asher's model to the data we get something very different.

Auto regressive model fit to simulated data

Auto regressive model fit to simulated data

His model finds a non-significant p-value of .49. We can also see the regression to the mean in his model with this simulated data.

Regression to the mean in simulated data

So, despite the fact that, after you adjust for the trend line, our simulated data is generating independent draws from a normal distribution, we see a negative auto-correlation in Mr. Asher's model due to model misspecification.

Final Thoughts

What we have shown is that the model proposed by Mr. Asher to "disprove" the theory of global warming is likely misspecified. It fails to to detect the highly significant trend that was present in our simulated data. Furthermore, if he is to call himself a statistician, he should have known exactly what was going on because regression to the mean is a fundamental 100 year old concept.



The data/code to reproduce this analysis are available here.




Filed under: R 18 Comments

How the Democrats may have won the House, but lost the seats


The 2012 election is over and in the books. A few very close races remain to be officially decided, but for the most part everything has settled down over the last week. By all accounts it was a very good night for the Democrats, with wins in the presidency, senate and state houses. They also performed better than expected in the House of Representatives. If we take the results as they stand now, the Democrats will have 202 seats v.s. the Republican's 233 seats, a pick-up of 9 relative to the 2010 election. This got me thinking about how the Republicans could have maintained 54% of the house seats during a Democratic wave election.

Since I'm a statistician, the obvious answer was to go get the data and find out. Over the last few days, I was able to obtain unofficial vote counts for each congressional race except 3. The results from two unopposed Democrats in Massachusetts' 1st and 2nd district, and one unopposed Republican from Kansas' 1st district will not be available until the final canvas is done in December. Not including these districts 55,709,796 votes were cast for Democrats, v.s. 55,805,487 for Republicans, a difference of less than 100,000 votes.

Both of the candidates from Massachusetts also ran unopposed in 2008, where they garnered 234,369 and 227,619 votes respectively, which seems like a good estimate for what we can expect this year. In 2008 Kansas' 1st district had a total of 262,027 votes cast. We will assign all of them to the Republican candidate as an (over) estimate, as the old 2002 district lines are similar to their 2012 placements. Using these numbers we get estimated vote counts of 56,171,784 for the Democrats and 56,067,514 for the Republicans, leading to a 100,000 vote advantage for the Democrats. Either way, this election was a real squeaker in terms of popular vote, which in percentage terms is 50.046% to 49.954%. Until the full official results are released, I would classify this as "too close to call," but find a Democrat popular vote victory likely.

Nate Silver recently looked at the differences in the number of votes cast in 2008 v.s. those already reported in 2012, and found that California has reported 3.4 million fewer votes this election than 2008, suggesting that millions of mail-in ballots remain waiting to be certified. Somewhat similarly, New York reported 1.5 million fewer votes, though this may be due (at least in part) to the effects of hurricane Sandy. These were the only two states with differences > 1 million, and they both lean heavily Democratic, which may well push the popular vote totals for the Democrats into "clear winner" territory.

In a certain sense, the winner of the popular vote is irrelevant, as the only thing that matters in terms of power is the number of seats held by a party. But if the Democrats win, then they can reasonably claim that they (even as the minority party) are the ones with the mandate to enact their agenda.

Even though the popular vote is razor close in the current unofficial results, the distribution of seats is certainly not. We shouldn't expect there to be a perfect correspondence between the popular vote and seats, because who wins the seats depends on how the population is divided up into congressional districts. As a simplified example consider three districts, each with 100,000 voters. District 1 is heavily Democratic, with 100% of votes going to the democratic candidate, while districts 2 and 3 are swing districts where 51% of the vote goes to the Republicans. In this simplified example, the Democrats would win nearly 2/3 of the popular vote, but only 1/3 of the seats.

So, if we expect there to be a discrepancy between popular vote and seat totals, the question then becomes: Is this election unusual? To answer this, we can look at the official historical record going back to 1942 which gives us the following:

Relationship between popular vote, and share of congressional seats (1942-2012)

In almost every election in the last 70 years, the party with the popular vote victory also won the majority of seats. The only exception to this was 1996, where the Democrats won the popular vote, but failed to attain a majority in congress. If 2012 turns out to be a popular vote victory for the Democrats, then their seat deficit would be unheard of in living memory. On the other hand, there was a very close election in 1950, where the Republicans lost by a margin of just 0.04%, but ended up with only 199 seats to the Democrats 234.

Redistricting, Gerrymandering, oh my!

Whether or not the official vote tally goes to the Democrats, the 2012 House of Representatives electoral landscape is unusual. The seat totals clearly favor the Republicans, but why? Some have noted that the congressional district lines were redrawn following the 2010 Census, a process that in many states is controlled by the state legislature, and that many of these lines seem to be constructed to maximize the political advantage of one party or the other. I was initially skeptical of these claims. After all, the two years most comparable to 2012 are 1996 and 1950, neither of which are particularly near a redistricting period. As is often the case, an examination of the data revealed evidence that was counter to my expectations.

First, let us explore how one would draw lines that maximize party advantage. Remember in our simple three district example, the advantage was gained by bunching all of the Democratic voters into one district, and maintaining slim advantages in the other two. So, the rule is to spread out your own vote into many districts, and consolidate your opponents voters into just a few. This can be a difficult geometric "problem" to solve, but an easy way to think about one possible route toward a solution is to focus on the swing districts. If a district is very close, you will want to alter its location so that it has just enough more Republican votes to make it a safe bet for your party. Ideally you will want to take these Republican votes from a safe Democratic district, making it even more Democratic. What you can then expect when you look at the distribution of districts is a clump of very safe Democratic districts, a clump of safe, but not overwhelmingly safe, Republican districts, and very few toss-up districts. This is what we call a bimodal distribution, and the process of redistricting with a focus on political gain is known as gerrymandering. There is nothing particularly Republican about it. Both political parties engage in gerrymandering, the difference being whether or not they control the redistricting process, and to what degree are they willing to contort the shape of their districts.

Authority to draw lines # of Congressional districts
Democratic party 44
Republican party 174
Split between parties 83
Independent Commission 88
Only one district in the state 7
Non-partisan legislature 3
Temporary court drawn 36

The Republican party was in control of the redistricting of 174 districts, v.s. just 44 for the Democrats. Just to give a specific example, Ohio had a Republican controlled redistricting process, which though wildly successful for Republicans, lead to torturously shaped districts. Republicans won 52.5% of the vote in Ohio, but obtained a staggering 75% of the seats.

But that is just one state, what about the rest of the districts? If we plot the proportion of people in each district that go to the Democrats, broken down by who was in control of redistricting we see clear evidence of clumping into safe, but not overwhelmingly safe territory for the redistricting party.

Share of vote by who controlled redistricting

The way to read this plot is that each point represents a district, with its position being the proportion of people in the district who voted for the Democratic candidate. The lines represent a smoothed estimate of the distribution, with wider parts indicating increased density. Among the districts where Republicans had control, we see a characteristic bimodal distribution, with many districts chosen to be relatively safe Republican strongholds, with a smaller group of heavily Democratic districts. The same trend may be present in the Democratically controlled districts, but there are too few of them to see the trend clearly. Alternatively, in districts where the decision making is split between parties, or delegated to an independent group, we see a more natural distribution of vote shares, with a good number of toss-up districts.

But can we prove that it swung the election?

To parse out the effect of redistricting on the election results in a non-descriptive sense, it is necessary to construct (a simple idealized) model of how redistricting affects the probability of winning a district. To do this, we assume that the probability of winning a district within a state is related to the partisanship of the state (i.e. we expect a higher proportion of Republican seats in Alabama as compared to New York), which we will measure by the share of the vote that Romney won in the state. Second, control of redistricting (either Democratic, Republican, or Independent/Split) yields an increase (or decrease) in the odds of a win. We slap these two together into a standard statistical model known as a logistic regression, which yields the following:

Variable Df Chi-squared p-value
Romney vote 1 31.4 <0.001
Redistricting control 2 9.1   0.010
Residuals 385

This table tells us two important things through its p-values. With p-values, the closer to 0, the more significant the relationship. First, that the partisanship of a district matters, because the proportion voting for Romney is very significantly related to the probability of congressional control. Secondly, control of redistricting matters. If the redistricting process were fair, we would see the magnitude of the trends that we see in this data only once every thousand years (10 years between redistricting / 0.010 ). This gives us a high degree of confidence that political concerns are the driving force in redistricting when politicians are put in charge of it (shocking, I know).

This effect is relatively large. Consider an idealized swing state, where Romney won exactly 50% of the vote. Our model estimates the proportion of the congressional delegation who are Democrats as:

Estimates of the average proportion of a congressional delegation from a true swing state who are Democrats based on a logistic regression

Here the model estimates that districts from an independent/split source are roughly unbiased with an estimated proportion of seats of 48%. If the Republicans control the process then we would expect only 31% of the delegation to be Democrats. This is slightly above what we observed in Ohio, where only 25% of the delegation were Democrats, despite Romney's percentage of the vote being nearly 50%. With the Democrats in control we see a smaller bias than the Republicans with an estimated 56% belonging to the Democratic party, but our confidence in this estimate (denoted by the red dotted lines) is small due to the small number of redistricting opportunities that they had.

A perfect world

What would have happened if the entire redistricting process was controlled by independent or bipartisan agreement? Well, our simple model can tell us something about that. By counterfactually assuming that all redistricting was done independently or by split legislature, it estimates that the proportion of Democrats in the current congress would be, on average, 52%. If Democrats got to pick all the lines it would be 59%, and under complete Republican control it would be just 37%.


So what lessons can we take away from this analysis? Control of the redistricting process is a powerful tool for those who would use it as a tool. The model estimates that who controls the process could yield  17% point swings (56%-39%) in a typical swing state. This is consistent both with the results from the general election as a whole, and Ohio in particular. While I would love to see the implementation of an algorithmic solution to the problem of district formation (if only to hear the pundits trying to say minimum isoperimetric quotient), it appears that independent commissions, or even bipartisan agreement within a legislative body are sufficient to have fair lines. As of now, 6 states decide their districts based on an independent commission, and it is hard to think of an honest argument why every state should not adopt this model. Economics teaches us that incentives matter, and if you give politicians the incentive to bias districts in their favor, it is a safe bet that they will do so.



Fellows Statistics provides experienced, professional statistical advice and analysis for the corporate and academic worlds.


Filed under: R 2 Comments

wordcloud makes words less cloudy


An update to the wordcloud package (2.2) has been released to CRAN. It includes a number of improvements to the basic wordcloud. Notably that you may now pass it text and Corpus objects directly. as in:

wordcloud("May our children and our children's children to a 
thousand generations, continue to enjoy the benefits conferred 
upon us by a united country, and have cause yet to rejoice under 
those glorious institutions bequeathed us by Washington and his 

SOTU <- tm_map(SOTU,function(x)removeWords(tolower(x),stopwords()))
wordcloud(SOTU, colors=brewer.pal(6,"Dark2"),random.order=FALSE)

This bigest improvement in this version though is a way to make your text plots more readable. A very common type of plot is a scatterplot, where instead of plotting points, case labels are plotted. This is accomplished with the text function in base R. Here is a simple artificial example:

states <- c('Alabama', 'Alaska', 'Arizona', 'Arkansas', 
	'California', 'Colorado', 'Connecticut', 'Delaware', 
	'Florida', 'Georgia', 'Hawaii', 'Idaho', 'Illinois',
	'Indiana', 'Iowa', 'Kansas', 'Kentucky', 'Louisiana',
	'Maine', 'Maryland', 'Massachusetts', 'Michigan', 'Minnesota',
	'Mississippi', 'Missouri', 'Montana', 'Nebraska', 'Nevada', 
	'New Hampshire', 'New Jersey', 'New Mexico', 'New York', 
	'North Carolina', 'North Dakota', 'Ohio', 'Oklahoma', 'Oregon', 
	'Pennsylvania', 'Rhode Island', 'South Carolina', 'South Dakota', 
	'Tennessee', 'Texas', 'Utah', 'Vermont', 'Virginia', 'Washington', 
	'West Virginia', 'Wisconsin', 'Wyoming')

loc <- rmvnorm(50,c(0,0),matrix(c(1,.7,.7,1),ncol=2))


Notice how many of the state names are unreadable due to overplotting, giving the scatter plot a cloudy appearance. The textplot function in wordcloud lets us plot the text without any of the words overlapping.


A big improvement! The only thing still hurting the plot is the fact that some of the states are only partially visible in the plot. This can be fixed by setting x and y limits, whch will cause the layout algorithm to stay in bounds.

mx <- apply(loc,2,max)
mn <- apply(loc,2,min)

Another great thing with this release is that the layout algorithm has been exposed so you can create your own beautiful custom plots. Just pass your desired coordinates (and word sizes) to wordlayout, and it will return bounding boxes close to the originals, but with no overlapping.

nc <- wordlayout(loc[,1],loc[,2],states,cex=50:1/20)
text(nc[,1] + .5*nc[,3],nc[,2]+.5*nc[,4],states,cex=50:1/20)

okay, so this one wasn't very creative, but it begs for some further thought. Now we have word clouds where not only the size can mean something, but also the x/y position (roughly) and color. Done right, this could add whole new layer of statistical richness to the visually pleasing but statistically shallow standard wordcloud.


Filed under: R, wordcloud 7 Comments