aaronjg

17 ms·

91.
by aaronjg 14y ago
The problem is that 'me not wanting to ever see your ad again' is NOT quantifiable. Marketers can see the positive impact of QR codes but not the negative. Thus by their metrics the more prominent the QR code, the more people who swipe it a
92.
by aaronjg 14y ago
Option 1 will NOT give you the correct answer. You CANNOT use confidence intervals as a stopping criteria. If you do this, you end up running many tests, and then you need to apply a multiple test correction to account for this. Otherwise y
93.
by aaronjg 14y ago
Rather than try to mine historical data, run an experiment to pit UCB against Neyman-Pearson inference. For some A/B tests, split the users into two groups. Treatment A is A/B testing, treatment B is UCB. In A/B testing, follow appropriate
94.
by aaronjg 14y ago
There are appropriate solutions to the multi-armed bandit problem, and a wealth of literature out there, however this is not one of those solutions. Here's a simple thought experiment to show that this will not 'beat A/B testing every time.
95.
A Crossword Puzzle That Doubles As A QR Code(fastcodesign.com)
5 points by aaronjg 14y ago|0 comments
96.
by aaronjg 14y ago
Y-axis need to start at zero when the data being presented is a bar graph and the bar is filled in. If it is a scatter plot with points being plotted, then it's generally not misleading to shift the axis.
97.
How VCs Calculate Valuation Differently from Founders(bothsidesofthetable.com)
2 points by aaronjg 14y ago|0 comments
98.
Google gets license to test drive autonomous cars on Nevada roads(arstechnica.com)
197 points by aaronjg 14y ago|164 comments
99.
by aaronjg 14y ago
Somewhat minor point, but I take issue with the CLV calculations. Using average retention rates can significantly undervalue your customers [1]. [1] http://blog.custora.com/2011/08/why-average-retention-rates-...
100.
by aaronjg 14y ago
Just sent you a copy of the paper. If you plan to use the result 'forever' then theoretically you would be willing to sacrifice a huge (infinite) amount of suboptimal performance now, so that you get the correct answer in the for when you d
101.
by aaronjg 14y ago
What libaries are you currently using where you would like to have things like this?
102.
by aaronjg 14y ago
In the clinical trial literature Anscombe's approach is considered Bayesian, and Armitage is frequentist. From Armitage's 1963 response to Anscombe's paper: 'Anscombe takes the Bayesian view that inferences should be made in terms of the li
103.
by aaronjg 14y ago
Anscombe talks about this and proposes two solutions: One is to estimate it based on the number of daily visitors your site gets, and then estimate how long you will run the winning alternative in the campaign. He also proposes: 'perhaps k
104.
by aaronjg 14y ago
Noel Welsh at untyped has been working on a cool implementation of this adaptive design. Check it out http://untyped.com/untyping/2011/02/11/stop-ab-testing-and-m... And the HN discussion: http://news.ycombinator.com/item?id=3867380
105.
How to Test a Random Number Generator(johndcook.com)
2 points by aaronjg 14y ago|0 comments
106.
by aaronjg 14y ago
I'm very impressed with their openness and in describing the process. They share revenue projections, their cap table and pitch deck. I'm also impressed with their non-standard move: They 'each sacrificed shares we owned to give back to ea
107.
Run Your Startup Like a 1940s Hardware Store(patrickleahy.me)
11 points by aaronjg 14y ago|0 comments
108.
by aaronjg 14y ago
I'd love to see these job queuing platforms have better support for high performance computing (HPC). Currently there are two paradigms of queuing systems. Things like PBS/Torque and Sun/Oracle/Univa Grid engine which work very well for sma
109.
by aaronjg 14y ago
I wrote a perl script to pull data from kickstarter. The backers page on the kickstarter page lists all of the backers, and the date that they made their pledge. I wrote a script to pull these dates. Data are only available at day level gra
110.
The A/B Test: Inside the Technology That’s Changing the Rules of Business(wired.com)
13 points by aaronjg 14y ago|2 comments
111.
by aaronjg 15y ago
I find the example benchmark to be rather ridiculous. Recursion is slow in R because R is an interpreted language, and does not implement tail-call optimization. If I wanted to implement the Fibonacci numbers in R I would do it like this: x
112.
by aaronjg 15y ago
I've run into this problem often when dealing with log probabilities in maximum likelihood estimation where the optimizer can try some numbers that are way off from the optimal value and you still want to return an accurate number. There is
113.
by aaronjg 15y ago
It sounds like you have done a lot of great customizations to your emacs. Care to share your .emacs configuration?
114.
by aaronjg 15y ago
That's a really cool feature. It would be interesting to see if users in different locales respond to differently to variations. For example, in an A/B test, perhaps US visitors respond better to variation B, and Canadian users respond bett
115.
by aaronjg 15y ago
Great introduction. I would love to see this as a PDF for easy reference.
116.
by aaronjg 15y ago
We often struggle writing for both audiences, and your feedback is well taken. Here's a brief rundown of the math, more details can be found in the papers linked below [1,2]. We assume a latent attrition model, that is customers purchase wi
117.
by aaronjg 15y ago
ced: We do some post-hoc analysis to find differences on whatever dimensions our clients give us. We have found that in general, the largest effect is the month of acquisition, and so this is the only factor that we include in the model rig
118.
by aaronjg 15y ago
We've considered hierarchical modeling, but concluded that there were no real gains. We have enough data from each of our clients to identify the parameters of the model. We will probably add more hierarchical modeling as we improve predic
119.
by aaronjg 15y ago
Indeed. Here is the original paper if you are interested in more details. http://www.jstor.org/pss/2631608
120.
by aaronjg 15y ago
We use the gamma prior and assume that shape parameter remains constant across time, and the scale parameter varies month to month. For each new month there are only two parameters that need to be estimated, one for the attrition rate and o

More ›