Estimates the effect of the degree on the rate of sampling. Also known as the "coefficient of discoverability" in the oil-discovery literature [2]. Formally, we estimate θ and β_0 assuming that β_k:= β_0 * k^θ.
1 |
rds.object |
A |
bin |
Bin degree counts. See Note. |
robust |
Should β_0 and θ be recovered from β_k using a robust method (default) or not. |
A list including the following components:
log.beta_0 |
The log of β_0 in β_k:= β_0 * k^θ. |
theta |
θ in β_k:= β_0 * k^θ. |
If degree counts have been binned by initializeRdsObject
(for variance reduction), the same value has to be supplied to getTheta
for correct estimation.
[1] Berchenko, Yakir, Jonathan Rosenblatt, and Simon D. W. Frost. "Modeling and Analyzing Respondent Driven Sampling as a Counting Process."" arXiv:1304.3505. HTTP://arXiv.org/abs/1304.3505. [2] Bloomfield, P., K.S. Deffeyes, B. Silverman, G.S. Watson, Y. Benjamini, and R.A. Stine. Volume and Area of Oil Fields and Their Impact on Order of Discovery, 1980. http://www.osti.gov/scitech/servlets/purl/6037591.
Estimate.b.k
,
initializeRdsObject
,
makeRdsSample
Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.
Please suggest features or report bugs with the GitHub issue tracker.
All documentation is copyright its authors; we didn't write any of that.