| Tag |
First Indicator |
Second Indicator |
Subfields |
| LEADER |
00000ctm a22000005a 4500 |
| 001 |
in00001422321 |
| 005 |
20190325093517.0 |
| 008 |
980708s1997 xx a b 000 0 eng d |
| 035 |
|
|
|a (OCoLC)39450058
|
| 035 |
|
|
|9 AHR2007AM
|
| 037 |
|
|
|a 98-15820
|b UMI
|
| 040 |
|
|
|a TXA
|c TXA
|d UtOrBLW
|
| 049 |
|
|
|a TXAM
|a TXAR
|
| 099 |
|
|
|a 1997
|a Dissertation
|a P3752
|
| 100 |
1 |
|
|a Park, Eun Sug ,
|d 1967-
|
| 245 |
1 |
0 |
|a Multivariate receptor modeling from a statistical science viewpoint /
|c by Eun Sug Park.
|
| 264 |
|
1 |
|a [Place of publication not identified] :
|b [publisher not identified] ;
|c 1997.
|
| 300 |
|
|
|a xi, 153 leaves :
|b illustrations ;
|c 28 cm.
|
| 336 |
|
|
|a text
|b txt
|2 rdacontent
|
| 337 |
|
|
|a unmediated
|b n
|2 rdamedia
|
| 338 |
|
|
|a volume
|b nc
|2 rdacarrier
|
| 504 |
|
|
|a Includes bibliographical references: pages 143-147.
|
| 500 |
|
|
|a Vita.
|
| 502 |
|
|
|b Ph. D.
|c Texas A&M University
|d 1997.
|
| 500 |
|
|
|a "Major Subject: Statistics".
|
| 530 |
|
|
|a Issued also on microfiche from University Microfilms Inc.
|
| 520 |
|
|
|a Receptor modeling is a collection of methods used to model
|
| 520 |
|
|
|a the air pollution data. Although the main interest is
|
| 520 |
|
|
|a identifying pollution sources, the determination of the
|
| 520 |
|
|
|a number of major sources is the first problem that we
|
| 520 |
|
|
|a should overcome. The NUMFACT algorithm is a new method of
|
| 520 |
|
|
|a determining the number of underlying factors in a
|
| 520 |
|
|
|a multivariate system. Although NUMFACT has not been
|
| 520 |
|
|
|a previously described, it has been used in high profile
|
| 520 |
|
|
|a studies. The asymptotic distribution of NUMFACT
|
| 520 |
|
|
|a statistics and their associated cutoff points are derived.
|
| 520 |
|
|
|a Modified NUMFACT statistics and related statistics, W, are
|
| 520 |
|
|
|a also presented with a new decision rule for determining
|
| 520 |
|
|
|a the number of sources. The simulation study shows that
|
| 520 |
|
|
|a for the lognormal error case, they are superior to the
|
| 520 |
|
|
|a traditional methods such as Bartlett's statistic or the
|
| 520 |
|
|
|a rule of thumb methods, and for the normal error case, they
|
| 520 |
|
|
|a are competitive with the best of the traditional methods.
|
| 520 |
|
|
|a Many air pollution datasets typically consist of the
|
| 520 |
|
|
|a measurements on fifty or sixty variables, and it is often
|
| 520 |
|
|
|a too large to handle all at once. We develop two new
|
| 520 |
|
|
|a algorithms, SPECIESA and SPECIESB, for choosing the
|
| 520 |
|
|
|a species used in final model fitting. The simulation
|
| 520 |
|
|
|a results show that these algorithms work very well in
|
| 520 |
|
|
|a choosing the species generated by the sources not by the
|
| 520 |
|
|
|a measurement errors. Estimating the source profiles and
|
| 520 |
|
|
|a their contributions are our primary concerns in receptor
|
| 520 |
|
|
|a modeling. We take the constrained nonlinear least squares
|
| 520 |
|
|
|a approach to provide those estimates under as little
|
| 520 |
|
|
|a assumptions as possible about the model parameters. The
|
| 520 |
|
|
|a assumptions that we make are necessary to get the model
|
| 520 |
|
|
|a identifiability. A set of algorithms, VERTEX, to find the
|
| 520 |
|
|
|a least squares solution under different versions of the
|
| 520 |
|
|
|a model is introduced. The resulting estimators are shown
|
| 520 |
|
|
|a to be consistent and asymptotically normal under
|
| 520 |
|
|
|a appropriate identifiability conditions. The proposed
|
| 520 |
|
|
|a methods are applied to simulated data and real air
|
| 520 |
|
|
|a pollution data.
|
| 650 |
|
4 |
|a Major statistics.
|
| 856 |
4 |
1 |
|u http://proxy.library.tamu.edu/login?url=http://proquest.umi.com/pqdweb?did=736824721&sid=1&Fmt=2&clientId=2945&RQT=309&VName=PQD
|t 0
|
| 948 |
|
|
|a c:en
|
| 999 |
|
|
|a MARS
|
| 999 |
f |
f |
|s fc23b7d0-2bfc-3942-a452-9c54f91a7331
|i 85a9ee00-1d92-3566-b375-6dcd062fcdd7
|t 0
|
| 952 |
f |
f |
|p noncirc
|a Texas A&M University
|b College Station
|c Cushing Memorial Library & Archives
|s cush tdrm
|d Cushing: Theses & Dissertations Microforms (Does not check out)
|t 0
|e 1997 Dissertation P3752
|h Other scheme
|i unmediated -- volume
|
| 952 |
f |
f |
|a Texas A&M University
|b College Station
|c Electronic Resources
|s www_evans
|d Available Online
|t 0
|e 1997 Dissertation P3752
|h Other scheme
|
| 998 |
f |
f |
|a 1997 Dissertation P3752
|t 0
|l Cushing: Theses & Dissertations Microforms (Does not check out)
|
| 998 |
f |
f |
|a 1997 Dissertation P3752
|t 0
|l Available Online
|