EXECUTIVE SUMMARY:   
   - We use a state-by-state list of number of missing people in 2019 to   
     determine which types of UFO activity appear to closely correlate.   
   - We also check some other interesting state demographics along with   
     the UFO activity.   
   - It seems some UFO types are related to increased number of missing   
     people; other types are robustly related to lower levels of missing   
     persons.   
   - At the same type HH firearms do not seem to reduce missing persons;   
     in some cases they may act as an "attractant". Perhaps those states   
     with a history of missing people are the ones that arm up.   
   - The number of Army bases in a state seems to have no effect on   
     missing persons related to some types of UFO activity. But other   
     types of UFO activity in the presence of Army bases seems to be   
     reduced.   
   - The presence of AFB have a similar spread as Army bases. Except some   
     types of UFO activity in the presence of AFB is seen to increase for   
     reasons unknown but maybe suspected. The surprise part is this is   
     2019 missing persons data, not 1960.   
      
      
   The data-scanning part of my s/w has managed to find a list of missing   
   people by US state. After downloading all the public FBI reports some   
   time back it seemed noone broke them down that way.   
      
   But someone must.   
      
   And this allows us to re-run some correlation studies to see if we can   
   find robust relationship between UFO activity over the US and volumes   
   of missing people -- even if it's only for the one year (2019).   
      
   So this won't be like most of the other studies where I want to ask   
   what data the s/w has on hand strongly correlates with this list of   
   missing person data; we want to just set up a model that involves some   
   of the things we've found in the past relate robustly to missing   
   persons and see what part UFO's play in that across the 50 states in   
   the dataset.   
      
   The summary information -- using the total UFO count between 2006 and   
   2020 per capita from NUFORC for each state -- looks like:   
      
   MODEL:   
   y = 1.28141*x + 22.951   
   beta in 1.28141 +- 0.50456   90% CI   
   alpha in 22.951 +- 11.7025   
   T-test: P(beta>0.000000) = 0.999952   
   Rank test: calculated Spearman corr = 0.507347   
   Critical Spearman = 0.432000 2-sided at 1%; reject H0:not_connected   
   r2 = 0.27868947   
      
   Data:   
   State		         #UFOs pc    #missing pc   model-predicted missing pc   
   TX                     7.82697        45.36      32.9806   
   New.York                  8.84026      30.6126       34.279   
   Louisiana                 8.99218      56.9505      34.4737*   
   Illinois                  10.4977      24.6501      36.4029   
   Maryland                  10.8218      30.4675      36.8182   
   Mississippi               11.0282      38.4316      37.0826   
   Alabama                   11.3193      41.9841      37.4557   
   Kansas                    12.0207      27.8194      38.3544   
   Iowa                      12.1643      27.5297      38.5384   
   Michigan                  12.2952      56.0338      38.7062   
   California                12.5943        54.49      39.0895   
   Georgia                   13.8034      24.4741      40.6388   
   Delaware                  14.8002      61.3151      41.9161   
   North.Carolina            15.2348      32.5606       42.473   
   Ohio                      15.3271      30.8264      42.5913   
   Massachusetts              15.601      18.5446      42.9423*   
   Nebraska                  15.8212      32.1698      43.2244   
   Virginia                   16.104      28.5101      43.5868   
   New.Jersey                16.2983      33.3779      43.8358   
   Kentucky                  16.4968       56.044      44.0901   
   Indiana                   16.6171      26.2853      44.2443   
   Pennsylvania              16.7936       31.322      44.4705   
   Oklahoma                   16.874      64.4281      44.5735   
   Tennessee                 17.2719      54.6945      45.0834   
   Florida                   17.4138      61.7623      45.2652   
   Arkansas                  17.4602      67.1546      45.3247*(1sd under)   
   South.Dakota               17.473      26.7919      45.3411   
   North.Dakota              18.4958      40.9551      46.6517   
   Missouri                  19.8893      51.9423      48.4373   
   West.Virginia             20.0637      65.0714      48.6608   
   Colorado                  20.5257      53.5134      49.2528   
   South.Carolina            21.4454      37.5806      50.4313   
   Minnesota                  22.406      32.6071      51.6622   
   Wisconsin                 22.8717      25.8172       52.259*   
   Nevada                    23.5225      66.4166      53.0929   
   Connecticut               25.0635      54.8611      55.0676   
   Utah                      27.0368      35.7153      57.5962*   
   Washington                27.6137      89.6748      58.3354*   
   New.Hampshire             30.0614      26.3038      61.4719*   
   Rhode.Island              30.2945      18.9341      61.7706**   
   Oregon                    31.2734      107.223       63.025**   
   Arizona                   31.6341      134.006      63.4872**(2sd under obs)   
   Hawaii                    33.5288      74.7414      65.9151   
   Wyoming                   34.1235      76.7778      66.6771   
   Vermont                   35.1414      86.2562      67.9815   
   New.Mexico                39.3265      68.5815      73.3443   
   Maine                     41.3743      81.2441      75.9684   
   Idaho                     44.7149      61.0298       80.249   
   Montana                   50.3413      68.7352      87.4588   
      
      
   Which (considering this is posting #3) renews the apparently robust   
   link between UFO activity/sightings and missing persons.  Previously   
   links were over time -- a number of years of FBI reports -- but this   
   one is for one year across the various states.   
      
   We see for each UFO sighted in a state about 1 person is reported   
   missing (1.3+-.5).   
      
   2 stats tests were done on the data. A T-test to make sure the   
   relevant \beta was too +ve to be just due to chance. The s/w   
   calculates there's less than 1 chance in 10 it could be so large just   
   by luck.   
      
   And a 2nd non-parametric rank test was done and also found the order   
   of states by UFO sightings and by missing person numbers was far too   
   similar to be just luck.   
      
   Together, we have something like 1 chance in 1000 or less the link is   
   just due to luck.   
      
   There is 99.9% something going on.   
      
   So far this is pretty much what we found before. We recall with the   
   FBI data we could also check each of the missing persons categories   
   and find the link was robust for some types of missing person   
   (e.g. juveniles) but not for others (dementia cases). And we also   
   checked the same pattern broadly was seen in UK and AUS missing data.   
      
   But we can now push on and check which UFO types are "most strongly   
   linked" and maybe add a couple other things to the mix like the number   
   of army and AF bases in each state, and maybe the% of   
   households with firearms. How do all these things together affect   
   missing person numbers in each state?   
      
   Most of what you expect is seen. There is one surprise. Or maybe not.   
      
   I've run each UFO type NUFORC keeps track of. The types are assigned   
   by the curators of the data rather than witnesses. We can thank them   
   for that. :)   
      
      
   [continued in next message]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)