Statistics Help @ Talk Stats Forum - Applied Statistics
http://www.talkstats.com/
Statistics in finance, economics, engineering. Actuarial science. Econometrics. Operations research.enThu, 11 Feb 2016 23:06:46 GMTvBulletin60http://www.talkstats.com/images/misc/rss.pngStatistics Help @ Talk Stats Forum - Applied Statistics
http://www.talkstats.com/
Which correlation analysis to use??
http://www.talkstats.com/showthread.php/63940-Which-correlation-analysis-to-use?goto=newpost
Thu, 11 Feb 2016 13:54:29 GMTI have 10 variables which are binary in nature. The two states are 1 and 0, where 1 refers to the negative effect of the variable and 0 refers to...I have 10 variables which are binary in nature. The two states are 1 and 0, where 1 refers to the negative effect of the variable and 0 refers to positive effect of the variable. The variables have an underlying normal distribution and are continuous.

I want to know which correlation analysis I can use to find correlation coefficients between these variables? Will it be Phi or tetrachoric?
]]>Applied Statisticsdhruv2phttp://www.talkstats.com/showthread.php/63940-Which-correlation-analysis-to-useInterest Rate Threshold Probability
http://www.talkstats.com/showthread.php/63936-Interest-Rate-Threshold-Probability?goto=newpost
Thu, 11 Feb 2016 02:39:56 GMTI am trying to complete a model that estimates the probability a bond yield hitting (or exceeding) a certain threshold, given historical rates going...I am trying to complete a model that estimates the probability a bond yield hitting (or exceeding) a certain threshold, given historical rates going back 20 years.

For example, the 10-year Treasury is currently at 1.71%. What is probability that the 10-year Treasury reaches 3.00% or greater in the next 3-year period?

There are many ways to look at this. For instance, maybe only looking at periods of rising rate environments (periods of economic expansions). However, at the most basic level, I am thinking of looking at just a general population of Treasury rates looking back twenty years, and refining the model beyond that at a later point.

Any ideas of how to attack this? Example models?

Thank you.
]]>Applied StatisticsOKUSAhttp://www.talkstats.com/showthread.php/63936-Interest-Rate-Threshold-ProbabilityMarketing - When to stop measuring
http://www.talkstats.com/showthread.php/63917-Marketing-When-to-stop-measuring?goto=newpost
Tue, 09 Feb 2016 12:41:18 GMTHello
I am running a marketing campaign with a treatment and a randomly selected control. To measure the success I am observing the number of...Hello

I am running a marketing campaign with a treatment and a randomly selected control. To measure the success I am observing the number of responses in both populations by week. I am measuring significance between the proportions using a Z-test. What I have seen so far is that the response proportions converge and are no longer significant by week 6 of the campaign. However if I add the total reponses between week 6 and 12 and recalculate then it is significant, although this feels wrong but I cant articulate why.

Please can someone help me explain if this is valid, and if not what would be an appropriate way of measuring the convergence between the population responses?

Thanks

Dave
]]>Applied StatisticsDavid Edwardshttp://www.talkstats.com/showthread.php/63917-Marketing-When-to-stop-measuring