You are on page 1of 8

MEMO

To: Kirk McClure


From: Qin He
Date: March 8, 2018
Subject: Influence of BMIR Program

According to the analysis, unsubsidized investment in the tracts which


participated in the BIMR program is $5,500 to $5,700 greater than unsubsidized
investment in the tracts that did not participate in the program. The spillover effects
the city hopes does exist.

Tracts participating in the BMIR program have higher unsubsidized investment

The city's 60 tracts census data were used to analyze whether participating in the
BMIR program stimulated the property owners’ enthusiasm for unsubsidized
investment in the repair, maintenance, and rehabilitation of their properties.

The average unsubsidized investment in this city is about 41,500. The average
date varies greatly with districts, from 14000 in the West to 60500 in the North. By
analyzing, the tracts participating in the BMIR program had average levels of
unsubsidized investment that were about 7300 higher than those did not participate.
But those data were not statistic significant to support this summary.

Darticipation Mean N Std. Deviation


Participate 45287.90 29 17428.346
Not Participate 37989.03 31 18002.797
Total 41516.82 60 17957.651
Table 1: Average unsubsidized investment separated by participation.
District Mean N Std. Deviation
North 60482.75 16 4444.769
East 48679.50 16 2725.877
South 39259.29 14 9690.060
West 13913.07 14 2521.324
Total 41516.82 60 17957.651
Table 2: Average unsubsidized investment separated by district

Table 1 and Table 2 describe the unsubsidized investment separated by district


and whether those tracts participated in the BMIR program.

Figure 1:distribution of unsubsidized investment in all tracts

Figure 1 illustrates the distribution of unsubsidized investment is not normal.

Analysis of the impact of the BMIR program on unsubsidized investment


According to one-way ANOVA analysis, the differences between participating
tracts and non-participating tracts are not statistically significant. The value of
significant probability of F is 0.116, much greater than 0.05.

ANOVA

df F Sig.
Between Groups 1 2.540 .116
Within Groups 58
Total 59
Table 3: ANOVA: Investment and BMIR program participation

Analysis of the impact of the district on unsubsidized investment


According to the one-way ANOVA analysis, the differences in four areas in this
city are statistic difference. The variation is statistically at better than the 0.005 level.
ANOVA

df F Sig.
Between Groups 3 188.893 .000
Within Groups 56
Total 59
Table 4: ANOVA: Investment and district

Contrast Coefficients

Contrast North East South West


1 1 0 0 -1
2 -1 -1 1 1
3 3 -1 -1 -1
4 0 1 0 -1
Table 5: Contrast pattern for ANOVA
Note:
⚫ The first contrast compares the highest investment district (North) to the lowest
investment district (West).
⚫ The second contrast compare the districts with below average investment (North
and East) with those above average investment (South and West).
⚫ The third contrast compares the highest district (North) with all other three
districts.
⚫ The forth contrast compares the two districts with the least variation (North and
South).

Contrast Tests:

Contrast t df Sig. (2-tailed)


invest Assume equal 1 23.021 56 .000
variances 2 -19.571 56 .000
3 16.433 56 .000
4 17.186 56 .000
Does not assume 1 35.835 24.273 .000
equal variances 2 -18.810 21.854 .000
3 18.388 29.950 .000
4 36.277 27.898 .000
Table 5: ANOVA: Investment and district: Contrast

Analysis of unsubsidized investment by BMIR program participation controlling


for district

Table 6 indicates that except East district, tracts participated in BMIR program
had higher level of investment.
According to the exception of east district, two-way ANOVA analysis should be
used in exploring the relationship between unsubsidized investment and statue of
BMIR program participation controlling for districts.
Table 7 indicates the differences between investment and participation are
statistically significant controlling for districts.
Table 8 indicates unsubsidized investment in the tracts which participated in the
BIMR program is about $5,500 greater than unsubsidized investment in the tracts that
did not participate in the program

Descriptive Statistics
Dependent Variable: invest
Contrast Results (K Matrix)
Dependent
Variable
part Simple Contrasta invest
Level 1 vs. Level 2 Contrast Estimate 5476.584
Hypothesized Value 0
Difference (Estimate - Hypothesized) 5476.584
Std. Error 1240.427
Sig. .000
95% Confidence Interval for Lower 2990.712
Difference Bound
Upper 7962.456
Bound
part district Mean Std. Deviation N
Participate North 63257.25 3488.795 8
East 48094.50 2968.430 8
South 47486.57 4508.933 7
West 15021.50 3455.191 6
Total 45287.90 17428.346 29
Not Participate North 57708.25 3545.060 8
East 49264.50 2515.651 8
South 31032.00 5017.407 7
West 13081.75 1198.201 8
Total 37989.03 18002.797 31
Table 6: Level of investment by participation and district

Tests of Between-Subjects Effects


Dependent Variable: invest
Source df F Sig.
Corrected Model 4 193.326 .000
Intercept 1 4298.282 .000
district 3 246.185 .000
part 1 19.493 .000
Error 55
Total 60
Corrected Total 59
Table 7: ANOVA: Investment and BMIR participation controlling for district
Table 8:Influence of additional investment by BMIR participation controlling for
district
Figure 2: Investment differences by districts and participation.

Analysis of unsubsidized investment by BMIR program participation controlling


for district and the interaction between district and participation

The pattern of investment across four districts also indicates the effects of
participation is different in different districts. It may be caused by the interaction
between district and participation.
Table 9 indicates the interaction between district and participation is statistically
significant.
Table 10 indicates unsubsidized investment in the tracts which participated in the
BIMR program is about $5,700 greater than unsubsidized investment in the tracts that
did not participate in the program

Tests of Between-Subjects Effects


Dependent Variable: invest
Source df F Sig.
Corrected Model 7 219.829 .000
Intercept 1 8194.439 .000
part 1 40.248 .000
district 3 473.550 .000
part * district 3 17.877 .000
Error 52
Total 60
Corrected Total 59

a. R Squared = .967 (Adjusted R Squared = .963)


Table 9: Influence of additional investment by BMIR participation controlling for
district and interaction between district and participation

Contrast Results (K Matrix)


Dependent Variable
part Simple Contrasta invest
Level 1 vs. Contrast Estimate 5693.330
Level 2 Hypothesized Value 0
Difference (Estimate - Hypothesized) 5693.330
Std. Error 897.412
Sig. .000
95% Confidence Interval for Lower 3892.541
Difference Bound
Upper 7494.120
Bound

a. Reference category = 2
Table 8:Influence of additional investment by BMIR participation controlling for
district and interaction between district and participation
Figure 2: Investment differences by districts, participation, and interaction between
district and participation.

Figure 10 indicates in North and south district, the effect of BMIR program are
positive, however, in East and West district, the spillover effect are small or did not
exist.

You might also like