As a South Asian Indian, I am puzzled by ethnicity questionnaires which never have an option that seems appropriate. "White" seems wrong as does "Caucasian", even if we are supposedly anthropologically related. "Asian" seems to refer to Asians except for persons of the Indian sub-continent. And "Indian" refers to American Indians.
Acquisitions provide liquidity for startups that hope to realize value only in the long-term. This is actually a good thing because it can actively encourage more ambitious startups.
I don't think Simpsons paradox plays any role here, if I understand this correctly.
In the first graph, the leading digit counts are not the things being summed up across all the regions. The weekly report counts are summed up across all regions and then the leading digits are counted.
Verizon's network is better than AT&T's for two fundamental reasons:
- It operates mostly on the 850MHz band whereas AT&T operates on both 850Mhz and 1900Mhz band. Lower frequencies penetrate buildings better than higher frequencies. Lower frequencies also have less interference with other electrical devices.
- Verizon uses CDMA technology which intrinsically makes more efficient use of the wireless spectrum than the GSM technology that AT&T uses.
Consequently, AT&T needs to expend more resources on constructing cell towers to achieve the same level of coverage as Verizon.
http://dabbledb.com/demo/ wowed me when I first saw it (still does). It was developed using the Seaside framework which has clean support for continuations.
I agree with the general idea that delegates from swing states should count for more.
The problem is your function doesn't really maximize the win probability either. I am not an expert here, but there may be a way to more accurately characterize the probability. Also, your function blows up when "average spread" tends to 0. An appropriate sigmoid transform should fix that.
Not sure why you need 3 points for 2D image registration. You need four variables here: two to determine the origin (x,y), one each for orientation and scale. You get these variables pinned with 2 points.
Sigh...