12.07.2015 Views

1 Studies in the History of Statistics and Probability ... - Sheynin, Oscar

1 Studies in the History of Statistics and Probability ... - Sheynin, Oscar

1 Studies in the History of Statistics and Probability ... - Sheynin, Oscar

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

But <strong>the</strong>n <strong>the</strong> newly appear<strong>in</strong>g peculiarities <strong>of</strong> that material will havenot been noticed.What k<strong>in</strong>d <strong>of</strong> peculiarities could happen? For example, on Fig. 2a<strong>and</strong> 2b a certa<strong>in</strong> non-zero number <strong>of</strong> observations is shown <strong>in</strong> <strong>the</strong>region µ* ≤ − 3. The probability <strong>of</strong> one observation be<strong>in</strong>g <strong>the</strong>re(assum<strong>in</strong>g that <strong>the</strong> Mendelian laws are valid) is 0.0014, <strong>and</strong>, <strong>of</strong> one out<strong>of</strong> approximately a hundred (to recall, <strong>the</strong> numbers <strong>of</strong> observationswere 98 <strong>and</strong> 123), about a hundred times higher; here, almost preciselyso. Thus, <strong>the</strong> probability <strong>of</strong> observations appear<strong>in</strong>g <strong>in</strong> that region <strong>in</strong>both series is about 0.14 2 ≈ 0.02, which means that a deviation fromnormality N(0, 1) is significant on <strong>the</strong> level ≈ 2%. So, are <strong>the</strong>Mendelian laws never<strong>the</strong>less wrong? Well, first <strong>of</strong> all, we have chosena test correspond<strong>in</strong>g to known data; second, a perfectly reasonableattitude does not mean dogmatically follow<strong>in</strong>g tests <strong>of</strong> significance. Areasonable answer apparently means that <strong>the</strong> bulk <strong>of</strong> observationsperfectly agrees with <strong>the</strong> Mendelian laws but sharp deviations perhapsdo occur. It can be supposed that a deficiency <strong>in</strong> <strong>the</strong> number <strong>of</strong>displayed recessive <strong>in</strong>dications has some biological sense (if,accord<strong>in</strong>g to a very simple explanation, <strong>the</strong>re exists a connection withsurvivorship).Incidentally, <strong>the</strong> above sufficiently illustrates <strong>the</strong> simple idea thattruth <strong>in</strong> science is established by <strong>the</strong> work <strong>of</strong> a number <strong>of</strong> generations<strong>and</strong> is not always atta<strong>in</strong>ed at each separate <strong>in</strong>vestigation.2.2. No one knows <strong>the</strong> hour ... The ancient say<strong>in</strong>g, No one knows<strong>the</strong> hour <strong>of</strong> his death, became somewhat shaken (certa<strong>in</strong>ly <strong>in</strong> <strong>the</strong>statistical ra<strong>the</strong>r than <strong>in</strong>dividual sense) after life tables have beencompiled <strong>and</strong> it occurred that <strong>the</strong> probability <strong>of</strong> liv<strong>in</strong>g up to a def<strong>in</strong>iteage, is subject to fluctuations (depend<strong>in</strong>g on <strong>the</strong> conditions <strong>of</strong> life),which are, however, not too essential. A fur<strong>the</strong>r step towards an<strong>in</strong>dividual forecast based on multivariate statistical analysis is partlymade <strong>and</strong> partly be<strong>in</strong>g made. I am describ<strong>in</strong>g one <strong>of</strong> <strong>the</strong> mostoutst<strong>and</strong><strong>in</strong>g contributions <strong>in</strong> this field, <strong>the</strong> so-called Fram<strong>in</strong>gham<strong>in</strong>vestigation (one <strong>of</strong> <strong>the</strong> pert<strong>in</strong>ent publications is Truett et al 1967).The cardiovascular diseases are known to be one <strong>of</strong> <strong>the</strong> centralproblems <strong>of</strong> modern medic<strong>in</strong>e. They are manifested <strong>in</strong> different ways;one <strong>of</strong> <strong>the</strong> most common k<strong>in</strong>d is <strong>the</strong> so-called ischemic heart disease(IHD). Accord<strong>in</strong>g to <strong>the</strong> classification adopted <strong>in</strong> <strong>the</strong> cited work, itcomprises cases <strong>of</strong> myocardial <strong>in</strong>farction, coronary <strong>in</strong>sufficiency,ang<strong>in</strong>a pectoris <strong>and</strong> deaths occasioned by disturbances <strong>of</strong> <strong>the</strong> coronaryblood circulation. We know well enough that <strong>the</strong> IHD <strong>of</strong>ten affectspeople yet be<strong>in</strong>g <strong>in</strong> <strong>the</strong> prime <strong>of</strong> creative power which makes <strong>the</strong>problem especially acute.There exist some ra<strong>the</strong>r vague ideas on <strong>the</strong> part played by <strong>the</strong> factors<strong>of</strong> modern <strong>in</strong>dustrialized life <strong>in</strong> <strong>the</strong> development <strong>of</strong> <strong>the</strong> IHD (littlephysical activity, nervous-emotional stress, irrational diet, etc) <strong>and</strong>also by <strong>the</strong> possible <strong>in</strong>fluence <strong>of</strong> genetic factors. These ideas arecerta<strong>in</strong>ly extremely important but we would like to have, <strong>in</strong> addition togeneral (but <strong>in</strong>sufficiently clear <strong>and</strong> <strong>in</strong>completely proven) ideas someamount <strong>of</strong> scientific (i. e. trustworthy) <strong>in</strong>formation.That, perhaps not cover<strong>in</strong>g <strong>the</strong> entire problem, would provide areliable foundation for some practical conclusions. Important is, for106

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!