# NEW STATISTICAL METHODS FOR CANCER TREATMENT

> **NIH NIH R01** · DANA-FARBER CANCER INST · 1989 · $107,002

## Abstract

Non-parametric statistical methods are commonly used to plan
studies and analyze biomedical data.  A key feature of these
methods is the use of asymptotic theory to derive the
approximate permutational distribution of the relevant test
statistic.  Inference based on such asympototic approximations is
not always satisfactory.  Many data sets do not satisfy the large
sample criteria that are needed for the asymptotic results to hold.
It is thus desirable to have the option to base inference on the
exact permutational distribution of the test statistic.  In the past,
this has not been a practical option because of the formidable
computational difficulties inherent in such an approach.  Recent
spectacular advances in computer technology, combined with the
availability of efficient numberical algorithms for exact
permutational inference, can change the situation.  The main goal
of this research project is to develop such numerical algorithms
for four commonly encountered biomedical problems; linear rank
tests, logistic regression, group sequential design, and contingency
tables.  An interdisciplinary approach will be taken.  Powerful
techniques of network optimization, recursion, MonteCarlo
sampling with variance reduction, and hashing, from the fields of
operations research and computer science, will play a prominent
role in all the algorithms.  Many new areas for methodological
research will be stimulated by the development of these
algorithms.  Some examples are, importance sampling as a new
way to quantify asymptotic rates of convergence, comparison of
exact and asymptotic methods of inference, the impact of time
trends on twosample linear rank tests, extending exact
regression to ordered categorical or censored survival data, and
exact power calculations for the Wilcoxon test with tied data.  All
these new directions will be investigated.

The results of this research will be useful to practicing
statisticians for planning and analyzing cancer trials, to applied
epidemiologists conducting casecontrol studies, and to
theoretical statisticians needing exact benchmarks against which
to compare their asymptotic results.

## Key facts

- **NIH application ID:** 3170932
- **Project number:** 5R01CA033019-08
- **Recipient organization:** DANA-FARBER CANCER INST
- **SAM.gov UEI:** DPMGH9MG1X67
- **Principal Investigator:** CYRUS R MEHTA
- **Activity code:** R01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 1989
- **Award amount:** $107,002
- **Award type:** 5
- **Project period:** 1982-07-15 → 1991-03-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/3170932

## Citation

> US National Institutes of Health, RePORTER application 3170932, NEW STATISTICAL METHODS FOR CANCER TREATMENT (5R01CA033019-08). Retrieved via MORVS 2026-10-25 from https://api.morvs.ai/grant/nih/3170932. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*