bkj123

Not sure what we can get from this article as there are so many variables... What "digital" spend was cut: paid search? online video? paid social media? brand vs. product oriented?

What was the impact of non-marketing drivers like distribution, price, promo, competition, and the economy?

What was the impact of (new) ad creative?

Maybe take it as they are under pressure to cut expenses and cutting marketing, digital and likely non-digital, certainly helps.

Thank you. very interesting. Helped me find other Google queries including SQLFiddle.

Is anyone doing this for analytics/data science/statistics or business intelligence/data warehousing?

I'm the same and did not discover this until about the 5th time I took the Myers-Briggs Personality Test. My initial tests had me almost exactly between extroversion and introversion.

On my last test, the facilitator explained the energy/stress aspect with a number of examples like your dinner party. She also mentioned people are naturally one or the other, not 50/50.

I realized that my natural inclination was introversion. I believe that what I perceived as my extroverted side was learned through a social upbringing to engage others, which is beneficial. There are a number of social/extroverted things that I sincerely enjoy but I walk away exhausted.

This is where a good analytical (as opposed to operational) data model comes into play. Something less normalized that is intuitive and efficient to use.

Regarding your last point... are you talking 'turnkey' as in something like 37 Signal's Basecamp?

Please suggest where I can find reliable facts and statistics regarding the conditions in US prisons including violence and rape. Thank you.

Regarding R's processing power, I haven't found it to be an issue. When building a model and testing, I use a sample of the data which is usually less than 100,000 observations. I use samples even when using a tool like SAS Enterprise Miner.

As far as scoring, I usually export using PMML and run it natively on the database. Makes for fast execution. PMML is available in R, RapidMiner, and other packages.