|
CONFIDENCE LIMITSName:
with \( \bar{x} \), s, n, and t denoting the sample mean, the sample standard deviation, the sample size, and the percent point function of the t distribution, respectively. This confidence interval is based on the assumption that the underlying data is approximately normally distributed. However, this confidence interval is fairly robust against non-normality unless the sample size is small or the departure from normality is severe (in particular, the data is not too skewed). For lognormally distributed data, the modified Cox method can be used to obtain a confidence interval for the mean. In this case, let Yi denote the log of Xi where X represents the original data. The confidence interval for the mean is then
with \( \bar{Y} \) and \( s \) denoting the mean and standard deviation of Y (i.e., the logged data) and t denotes the t percent point function. For confidence intervals, computing standard confidence limits for the logged data and then back transforming to obtain confidence limits for the original data does not generate accurate intervals. For sufficiently large samples (based on simulations, Olsson suggests sample sizes larger than 200), using the standard normal based confidence interval should give reasonable results. Using the BOOTSTRAP MEAN PLOT command is an alternative method to obtain the confidence interval for data that is not approximately normally distributed.
<SUBSET/EXCEPT/FOR qualification> where <y> is the response variable; <LOGNORMAL> is optional; and where the <SUBSET/EXCEPT/FOR qualification> is optional. If LOWER is specified, a one-sided lower confidence limit is returned. If UPPER is specified, a one-sided upper confidence limit is returned. If neither is specified, a two-sided limit is returned. If LOGNORMAL is given, the confidence interval based on the lognormal distribution will be used. This syntax supports matrix arguments for the response variable.
<y1> ... <yk> <SUBSET/EXCEPT/FOR qualification> where <y1> .... <yk> is a list of 1 to 30 response variables; <LOGNORMAL> is optional; and where the <SUBSET/EXCEPT/FOR qualification> is optional. This syntax will generate a confidence interval for each of the response variables. The word MULTIPLOT is optional. That is,
is equivalent to
If LOWER is specified, a one-sided lower confidence limit is returned. If UPPER is specified, a one-sided upper confidence limit is returned. If neither is specified, a two-sided limit is returned. If LOGNORMAL is given, the confidence interval based on the lognormal distribution will be used. This syntax supports matrix arguments for the response variables.
<y> <x1> ... <xk> <SUBSET/EXCEPT/FOR qualification> where <y> is the response variable; <x1> .... <xk> is a list of 1 to 6 group-id variables; <LOGNORMAL> is optional; and where the <SUBSET/EXCEPT/FOR qualification> is optional. This syntax performs a cross-tabulation of the <x1> ... <xk> and generates a confidence interval for each unique combination of the cross-tabulated values. For example, if X1 has 3 levels and X2 has 2 levels, six confidence intervals will be generated. If LOWER is specified, a one-sided lower confidence limit is returned. If UPPER is specified, a one-sided upper confidence limit is returned. If neither is specified, a two-sided limit is returned. If LOGNORMAL is given, the confidence interval based on the lognormal distribution will be used. This syntax does not support matrix arguments.
CONFIDENCE LIMITS Y1 SUBSET TAG > 2 MULTIPLE CONFIDENCE LIMITS Y1 TO Y5 REPLICATED CONFIDENCE LIMITS Y X
LET A = LOWER CONFIDENCE LIMIT Y The LET ALPHA = command is used to specify the significance level. In addition to the above LET command, built-in statistics are supported for about 20 different commands (enter HELP STATISTICS for details).
Olsson (2005), "Confidence Intervals for the Mean of a Log-Normal Distribution", Journal of Statistics Education, Vol. 13, No. 1.
2010/03: Support for MULTIPLE and REPLICATION options 2010/03: Support for matrix options 2017/07: Support for lognormal confidence limits SKIP 25 READ ZARR13.DAT Y SET WRITE DECIMALS 5 CONFIDENCE LIMITS YThe following output is generated. Confidence Limits for the Mean (Two-Sided) Response Variable: Y Summary Statistics: Number of Observations: 195 Sample Mean: 9.26146 Sample Standard Deviation: 0.02278 Sample Standard Deviation of the Mean: 0.00163 ----------------------------------------------------------------- Confidence t t-Value X Lower Upper Value (%) Value SD(Mean) Limit Limit ----------------------------------------------------------------- 50.000 0.675 0.00110 9.26035 9.26256 75.000 1.153 0.00188 9.25957 9.26334 90.000 1.652 0.00269 9.25876 9.26415 95.000 1.972 0.00321 9.25824 9.26467 99.000 2.601 0.00424 9.25721 9.26570 99.900 3.341 0.00545 9.25600 9.26691 99.990 3.973 0.00648 9.25497 9.26794 99.999 4.536 0.00740 9.25405 9.26886Program 2: SKIP 25 READ GEAR.DAT Y X SET WRITE DECIMALS 5 REPLICATED CONFIDENCE LIMITS Y XThe following output is generated. Confidence Limits for the Mean (Two-Sided) Response Variable: Y Factor Variable 1: X 1.00000 Summary Statistics: Number of Observations: 10 Sample Mean: 0.99800 Sample Standard Deviation: 0.00434 Sample Standard Deviation of the Mean: 0.00137 ----------------------------------------------------------------- Confidence t t-Value X Lower Upper Value (%) Value SD(Mean) Limit Limit ----------------------------------------------------------------- 50.000 0.702 0.00096 0.99703 0.99896 75.000 1.229 0.00169 0.99630 0.99969 90.000 1.833 0.00251 0.99548 1.00051 95.000 2.262 0.00310 0.99489 1.00110 99.000 3.249 0.00446 0.99353 1.00246 99.900 4.779 0.00656 0.99143 1.00456 99.990 6.584 0.00904 0.98895 1.00704 99.999 8.794 0.01208 0.98591 1.01008 Confidence Limits for the Mean (Two-Sided) Response Variable: Y Factor Variable 1: X 2.00000 Summary Statistics: Number of Observations: 10 Sample Mean: 0.99910 Sample Standard Deviation: 0.00521 Sample Standard Deviation of the Mean: 0.00164 ----------------------------------------------------------------- Confidence t t-Value X Lower Upper Value (%) Value SD(Mean) Limit Limit ----------------------------------------------------------------- 50.000 0.702 0.00115 0.99794 1.00025 75.000 1.229 0.00202 0.99707 1.00112 90.000 1.833 0.00302 0.99607 1.00212 95.000 2.262 0.00373 0.99536 1.00283 99.000 3.249 0.00536 0.99373 1.00446 99.900 4.779 0.00788 0.99121 1.00698 99.990 6.584 0.01086 0.98823 1.00996 99.999 8.794 0.01450 0.98459 1.01360 Confidence Limits for the Mean (Two-Sided) Response Variable: Y Factor Variable 1: X 3.00000 Summary Statistics: Number of Observations: 10 Sample Mean: 0.99540 Sample Standard Deviation: 0.00397 Sample Standard Deviation of the Mean: 0.00125 ----------------------------------------------------------------- Confidence t t-Value X Lower Upper Value (%) Value SD(Mean) Limit Limit ----------------------------------------------------------------- 50.000 0.702 0.00088 0.99451 0.99628 75.000 1.229 0.00154 0.99385 0.99694 90.000 1.833 0.00230 0.99309 0.99770 95.000 2.262 0.00284 0.99255 0.99824 99.000 3.249 0.00408 0.99131 0.99948 99.900 4.779 0.00601 0.98938 1.00141 99.990 6.584 0.00828 0.98711 1.00368 99.999 8.794 0.01106 0.98433 1.00646 Confidence Limits for the Mean (Two-Sided) Response Variable: Y Factor Variable 1: X 4.00000 Summary Statistics: Number of Observations: 10 Sample Mean: 0.99820 Sample Standard Deviation: 0.00385 Sample Standard Deviation of the Mean: 0.00121 ----------------------------------------------------------------- Confidence t t-Value X Lower Upper Value (%) Value SD(Mean) Limit Limit ----------------------------------------------------------------- 50.000 0.702 0.00085 0.99734 0.99905 75.000 1.229 0.00149 0.99670 0.99969 90.000 1.833 0.00223 0.99596 1.00043 95.000 2.262 0.00275 0.99544 1.00095 99.000 3.249 0.00395 0.99424 1.00215 99.900 4.779 0.00582 0.99237 1.00402 99.990 6.584 0.00802 0.99017 1.00622 99.999 8.794 0.01071 0.98748 1.00891 Confidence Limits for the Mean (Two-Sided) Response Variable: Y Factor Variable 1: X 5.00000 Summary Statistics: Number of Observations: 10 Sample Mean: 0.99190 Sample Standard Deviation: 0.00757 Sample Standard Deviation of the Mean: 0.00239 ----------------------------------------------------------------- Confidence t t-Value X Lower Upper Value (%) Value SD(Mean) Limit Limit ----------------------------------------------------------------- 50.000 0.702 0.00168 0.99021 0.99358 75.000 1.229 0.00294 0.98895 0.99484 90.000 1.833 0.00439 0.98750 0.99629 95.000 2.262 0.00542 0.98647 0.99732 99.000 3.249 0.00778 0.98411 0.99968 99.900 4.779 0.01145 0.98044 1.00335 99.990 6.584 0.01578 0.97611 1.00768 99.999 8.794 0.02107 0.97082 1.01297 Confidence Limits for the Mean (Two-Sided) Response Variable: Y Factor Variable 1: X 6.00000 Summary Statistics: Number of Observations: 10 Sample Mean: 0.99879 Sample Standard Deviation: 0.00988 Sample Standard Deviation of the Mean: 0.00312 ----------------------------------------------------------------- Confidence t t-Value X Lower Upper Value (%) Value SD(Mean) Limit Limit ----------------------------------------------------------------- 50.000 0.702 0.00219 0.99660 1.00099 75.000 1.229 0.00384 0.99495 1.00264 90.000 1.833 0.00573 0.99306 1.00453 95.000 2.262 0.00707 0.99172 1.00587 99.000 3.249 0.01015 0.98864 1.00895 99.900 4.779 0.01494 0.98385 1.01374 99.990 6.584 0.02058 0.97821 1.01938 99.999 8.794 0.02749 0.97130 1.02629 Confidence Limits for the Mean (Two-Sided) Response Variable: Y Factor Variable 1: X 7.00000 Summary Statistics: Number of Observations: 10 Sample Mean: 1.00150 Sample Standard Deviation: 0.00787 Sample Standard Deviation of the Mean: 0.00249 ----------------------------------------------------------------- Confidence t t-Value X Lower Upper Value (%) Value SD(Mean) Limit Limit ----------------------------------------------------------------- 50.000 0.702 0.00175 0.99974 1.00325 75.000 1.229 0.00306 0.99843 1.00456 90.000 1.833 0.00456 0.99693 1.00606 95.000 2.262 0.00563 0.99586 1.00713 99.000 3.249 0.00809 0.99340 1.00959 99.900 4.779 0.01190 0.98959 1.01340 99.990 6.584 0.01640 0.98509 1.01790 99.999 8.794 0.02190 0.97959 1.02340 Confidence Limits for the Mean (Two-Sided) Response Variable: Y Factor Variable 1: X 8.00000 Summary Statistics: Number of Observations: 10 Sample Mean: 1.00039 Sample Standard Deviation: 0.00362 Sample Standard Deviation of the Mean: 0.00114 ----------------------------------------------------------------- Confidence t t-Value X Lower Upper Value (%) Value SD(Mean) Limit Limit ----------------------------------------------------------------- 50.000 0.702 0.00080 0.99959 1.00120 75.000 1.229 0.00141 0.99898 1.00181 90.000 1.833 0.00210 0.99829 1.00250 95.000 2.262 0.00259 0.99780 1.00299 99.000 3.249 0.00372 0.99667 1.00412 99.900 4.779 0.00548 0.99491 1.00588 99.990 6.584 0.00755 0.99284 1.00795 99.999 8.794 0.01008 0.99031 1.01048 Confidence Limits for the Mean (Two-Sided) Response Variable: Y Factor Variable 1: X 9.00000 Summary Statistics: Number of Observations: 10 Sample Mean: 0.99829 Sample Standard Deviation: 0.00413 Sample Standard Deviation of the Mean: 0.00130 ----------------------------------------------------------------- Confidence t t-Value X Lower Upper Value (%) Value SD(Mean) Limit Limit ----------------------------------------------------------------- 50.000 0.702 0.00091 0.99738 0.99921 75.000 1.229 0.00160 0.99669 0.99990 90.000 1.833 0.00239 0.99590 1.00069 95.000 2.262 0.00296 0.99533 1.00126 99.000 3.249 0.00425 0.99404 1.00255 99.900 4.779 0.00625 0.99204 1.00455 99.990 6.584 0.00861 0.98968 1.00691 99.999 8.794 0.01150 0.98679 1.00980 Confidence Limits for the Mean (Two-Sided) Response Variable: Y Factor Variable 1: X 10.00000 Summary Statistics: Number of Observations: 10 Sample Mean: 0.99479 Sample Standard Deviation: 0.00532 Sample Standard Deviation of the Mean: 0.00168 ----------------------------------------------------------------- Confidence t t-Value X Lower Upper Value (%) Value SD(Mean) Limit Limit ----------------------------------------------------------------- 50.000 0.702 0.00118 0.99361 0.99598 75.000 1.229 0.00207 0.99272 0.99687 90.000 1.833 0.00308 0.99171 0.99788 95.000 2.262 0.00381 0.99098 0.99861 99.000 3.249 0.00547 0.98932 1.00027 99.900 4.779 0.00805 0.98674 1.00285 99.990 6.584 0.01109 0.98370 1.00589 99.999 8.794 0.01482 0.97997 1.00962Program 1: SKIP 25 READ LGN2.DAT Y SET WRITE DECIMALS 4 LOGNORMAL CONFIDENCE LIMITS YThe following output is generated. Two-Sided Confidence Limits for the Mean (Log-Normal, Modified Cox Method) Response Variable: Y Summary Statistics: Number of Observations: 40 Sample Mean (Raw Data): 274.96250 Sample Standard Deviation (Raw Data): 310.34271 Sample Mean (Log Data): 5.12720 Correction Term (s*2/2) 0.50439 Sample Standard Deviation (Log Data): 1.00438 Sample Standard Deviation of the Mean: 0.19562 ----------------------------------------------------------------- Confidence t t-Value X Lower Upper Value (%) Value SD(Mean) Limit Limit ----------------------------------------------------------------- 50.000 0.681 0.13318 244.30086 318.86400 75.000 1.168 0.22843 222.10554 350.72852 90.000 1.685 0.32959 200.73651 388.06468 95.000 2.023 0.39567 187.90017 414.57519 99.000 2.708 0.52971 164.32892 474.04163 99.900 3.558 0.69603 139.15044 559.81675 99.990 4.333 0.84756 119.58387 651.41519 99.999 5.071 0.99202 103.49946 752.64886
Date created: 04/15/2013 |
Last updated: 12/04/2023 Please email comments on this WWW page to alan.heckert@nist.gov. |