Method for analyzing demographic data

1. A computer-implemented method of selecting a geographic location having similarities to another geographic location in at least one user defined category, the method comprising the steps of:

a. selecting a plurality of geographic locations;

b. selecting a plurality of predefined characteristics applicable to the geographic locations;

c. defining a range of values for each characteristic, each value representating a degree to which a geographic location embodies the characteristic;

d. defining a data base of geographic locations and characteristics, and populating the data base with characteristic values for the geographic locations;

e. selecting a first geographic location from the data base; and,

f. comparing the first geographic location in the data base with other geographic locations in the data base and generating an ordered list of other geographic locations most similar to the first geographic location,

wherein the step of comparing the first geographic location in the data base with other geographic locations in the data base and generating an ordered list of other geographic locations most similar to the first geographic location comprises generating a dissimilarity matrix in which an overall dissimilarity value is assigned to each of the other geographic locations relative to the first location, and generating an ordered list of the other geographic locations therefrom, and

wherein the step of generating a dissimilarity matrix comprises the step of applying an extraction method to the raw data, the extraction method including a factor analysis to extract factors from the data base, and applying a squared Euclidian distance measure to the extracted factors.

2. A method according to claim 1 wherein the extraction method comprises a principal component factor analysis of a correlation matrix R is specified in terms of its eigenvalue-eigenvector pairs,

(

Ω

Ω

λ

1

,

λ

1

)

,

(

Ω

Ω

λ

2

,

λ

2

)

,

…

⁢

,

(

Ω

Ω

λ

N

,

λ

N

)

,

where

⁢

Ω

λ

1

⁢

≥

⁢

Ω

λ

2

≥

…

≥

Ω

λ

γ

;

where m>p is the number of common factors, and p is the total number of original variables; where the estimated specific variances are provided by the diagonal elements of the matrix R−{tilde over (L)}{tilde over (L)}′, such that

ψ

~

=

[

ψ

~

1

0

⋯

0

0

ψ

~

2

⋯

0

⋮

⋮

⋰

⋮

0

0

⋯

ψ

~

γ

]

⁢

⁢

with

⁢

⁢

ψ

~

i

=

R

ii

-

∑

j

=

1

m

⁢

l

~

ij

2

for ith variable, jth factor; where in communalities are estimated as {tilde over (h)} i 2 ={tilde over (l)} i1 2 +{tilde over (l)} i2 2 + . . . +{tilde over (l)} im 2 ; and where the number of factors extracted is set to capture 95% of the total variance contained in the original data.

3. A method according to claim 2 wherein the extracted factors for every characteristic are saved in the data base for each geographic location.

4. A method according to claim 3 wherein the saved factors for each geographic location are input to the formula

d

ij

=

∑

k

⁢

(

x

ik

-

x

jk

)

2

where d=distance, and x ik =value of factor k for census tract i, to calculate a dissimilarity matrix containing each geographic location.

5. A method according to claim 4 wherein an ordered list of geographic locations is displayed to a user, the order of the displayed geographic locations determined by the dissimilarity matrix.

6. A computer-implemented method of selecting a geographic location having similarities to another geographic location in at least one user defined category, the method comprising the steps of:

a. selecting a plurality of geographic locations;

b. selecting a plurality of predefined characteristics applicable to the geographic locations;

c. defining a range of values for each characteristic, each value representating a degree to which a geographic location embodies the characteristic;

d. defining a data base of geographic locations and characteristics, and populating the data base with characteristic values for the geographic locations;

e. selecting a first geographic location from the data base; and,

f. comparing the first geographic location in the data base with other geographic locations in the data base and generating an ordered list of other geographic locations most similar to the first geographic location,

wherein the step of comparing the first geographic location in the data base with other geographic locations in the data base and generating an ordered list of other geographic locations most similar to the first geographic location comprises generating a dissimilarity matrix in which an overall dissimilarity value is assigned to each of the other geographic locations relative to the first location, and generating an ordered list of the other geographic locations therefrom, and

wherein the step of generating a dissimilarity matrix comprises the step of applying an extraction method to the raw data that simultaneously eliminates multicolinearity from the raw data in the data base and standardizes the raw data in the data base.