HP (Hewlett-Packard) HP 33s ユーザーズマニュアル

ページ / 387
11–10
 Statistical Operations 
Normalizing Close, Large Numbers 
The calculator might be unable to correctly calculate the standard deviation and 
linear regression for a variable whose data values differ by a relatively small 
amount. To avoid this, normalize the data by entering each value as the difference 
from one central value (such as the mean). For normalized x–values, this difference 
must then be added back to the calculation of 
x
 and 
xˆ
, and 
yˆ
 and b must 
also be adjusted. For example, if your x–values were 7776999, 7777000, and 
7777001, you should enter the data as –1, 0, and 1; then add 7777000 back to 
x
 and 
xˆ
.
 For b, add back 7777000 
× m. To calculate 
yˆ
,
be sure to supply 
an x–value that is less 7777000. 
Similar inaccuracies can result if your and  values have greatly different 
magnitudes. Again, scaling the data can avoid this problem. 
Effect of Deleted Data 
Executing 
 does not delete any rounding errors that might have been 
generated in the statistics registers by the original data values. This difference is 
not serious unless the incorrect data have a magnitude that is enormous compared 
with the correct data; in such a case, it would be wise to clear and reenter all the 
data.
Summation Values and the Statistics Registers 
The statistics registers are six unique locations in memory that store the 
accumulation of the six summation values. 
Summation Statistics 
Pressing 
gives you access to the contents of the statistics registers: 
„
Press {Q} to recall the number of accumulated data sets. 
„
Press {º} to recall the sum of the x–values.  
„
Press {¸} to recall the sum of the y–values. 
„
Press {;º
}, {;¸
}, and {;º¸} to recall the sums of the squares and the sum 
of the products of the and — values that are of interest when performing 
other statistical calculations in addition to those provided by the calculator.