2009年から2年間、大学院で環境関連学を専攻するため、イギリスにやって来た私達の話 This blog is about us(U&I) coming to UK in 2009 for Environmental Study at master level for 2 years
by uk-env
1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31
Viva la statistics 統計はお好き?

I had been stragling with the assignment for Statistics & Quantitative Methods for the last three weeks. This was about the ecological analysis with statistics software "R" on the impacts of several factors such as coral reef size, fishery on the habitats of coral reef fishes - groupers. (See my paper from this)



The benefical thing for me is that I have learned various statistical tools (t-test, f-test, ks-test, ANOVA, Pearson correlation, OLS regression model, Poisson GLM etc etc), which can be used for the basic data analysis in the future.

The more interesting thing is to know the fact that different academic fields use different statistical tools and different analytical processes even though adopting the same stastical tool. For instance, Environmental Valuation, the other module, uses the econometric tool "LIMDEP" that economists and econometrists like. Also the professor of that Valuation module does not like AIC reduction model (backward step model to reduce non-significant variables) because of its little reliability but which was recommended in the above "Econological" stats course. It also depends on researchers' preferences though.

Thus, the Environmental Deaprtment of York University is considering to change the course structure from next year especially for Environmental Economics and Environmental Management and teach those students "consolidated" quantitative analytical methods from the above two varied academic fields (I don't how to manage it though). I actually enjoyed knowing such difference in statistical policy between two fields but that integration seems more efficient and reasonable.

On a side note, (according to my friend who has an expertise of insurance business,) insurers analyse the probability of risk occurrance from millions of risk factors that they collect around the world as sample data (e.g. for life insurance, our disease and accident records) with super complicated statistical and mathmatical models, in order to assess premium rate of insurance product. And mainly in the US, (notorious) investment companies have launched some insurace-related business since 2008 - with the analysis on the likelihood of life insurance claims as the same as insurers, these investors purchase life insurance policies from policyholders who need cash now, pay premiums for them, and receive the insurace instead of such policyholders when they go to heaven. Wow, what a clean business!

Anyway, the last few weeks was when I learned the "importance" of stats, and last night after the submission of the assignment, I enjoyed the party time with mates, as we say "screw stats" party. Right then, good-bye stats until my dissertation ;p

by uk-env | 2011-03-05 18:32 | 大学(York Uni)
<< Lovely Cornwall... Horse riding >>