#uploading and viewing of data
#df=read.csv('data.csv', header = TRUE)
#upload of libraries(all these libraries should firstly be installed at the packages section)
library(readr) # reading data of different types
library(ggplot2) # for visualization
library(plyr)
library(dplyr)
library(tidyr) #tidying up data
library(psych) #for modeling
library(caTools)
library(mice) #for cleaning
library(VIM) #for cleaning
library(viridis) #for additional coloring
library(knitr)
download.file("https://raw.githubusercontent.com/ashishsm1986/DATA698/main/Entrepreneurial_Acclimatization.csv",destfile = "Entrepreneurial_Acclimatization.csv",method = "curl")
df=read.csv('Entrepreneurial_Acclimatization.csv', header = TRUE)
#df=read.csv('Entrepreneurial Acclimatization.csv', header = TRUE)
head(df)The Dataset is a copy of dataset located here:- https://www.kaggle.com/manishkc06/startup-success-prediction as well as answers received from survey conducted for the following questions and they have been mapped into the file as additional columns:-
C1 Top management pushes for Research and Development, Technological acclimatization and innovations hence high entrepreneurial acclimatization?
C2 Projects are executed under stipulated budgets/timelines, with targets met and concerns addressed, hence bringing forth positive influence and satisfaction of different stakeholders.
C3 There is a fast adoption of new technological acclimatization and there is the tendency of using the most updated technology in performing tasks.Single choice.
C4 Top management has high rate of positive support and influence on projects hence high rate of success noted.Single choice.
Lets Dig into the data
## [1] 923 57The Data has 923 Rows and 57 Columns
| state_code | zip_code | funding_total_usd | category_code | status |
|---|---|---|---|---|
| CA | 92101 | 375000 | music | acquired |
| CA | 95032 | 40100000 | enterprise | acquired |
| CA | 92121 | 2600000 | web | acquired |
| CA | 95014 | 40000000 | software | acquired |
| CA | 94105 | 1300000 | games_video | closed |
##Data Preprocessing
#Deletion of the rows that have missing data cells
#checking for missing values and cleaning data
head(is.na(df))## ID state_code latitude longitude zip_code id city Unnamed..6 name
## [1,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [2,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [3,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [4,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [5,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [6,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## labels founded_at closed_at first_funding_at last_funding_at
## [1,] FALSE FALSE FALSE FALSE FALSE
## [2,] FALSE FALSE FALSE FALSE FALSE
## [3,] FALSE FALSE FALSE FALSE FALSE
## [4,] FALSE FALSE FALSE FALSE FALSE
## [5,] FALSE FALSE FALSE FALSE FALSE
## [6,] FALSE FALSE FALSE FALSE FALSE
## age_first_funding_year age_last_funding_year age_first_milestone_year
## [1,] FALSE FALSE FALSE
## [2,] FALSE FALSE FALSE
## [3,] FALSE FALSE FALSE
## [4,] FALSE FALSE FALSE
## [5,] FALSE FALSE FALSE
## [6,] FALSE FALSE FALSE
## age_last_milestone_year relationships funding_rounds funding_total_usd
## [1,] FALSE FALSE FALSE FALSE
## [2,] FALSE FALSE FALSE FALSE
## [3,] FALSE FALSE FALSE FALSE
## [4,] FALSE FALSE FALSE FALSE
## [5,] FALSE FALSE FALSE FALSE
## [6,] FALSE FALSE FALSE FALSE
## milestones state_code.1 is_CA is_NY is_MA is_TX is_otherstate
## [1,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [2,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [3,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [4,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [5,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [6,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## category_code is_software is_web is_mobile is_enterprise is_advertising
## [1,] FALSE FALSE FALSE FALSE FALSE FALSE
## [2,] FALSE FALSE FALSE FALSE FALSE FALSE
## [3,] FALSE FALSE FALSE FALSE FALSE FALSE
## [4,] FALSE FALSE FALSE FALSE FALSE FALSE
## [5,] FALSE FALSE FALSE FALSE FALSE FALSE
## [6,] FALSE FALSE FALSE FALSE FALSE FALSE
## is_gamesvideo is_ecommerce is_biotech is_consulting is_othercategory
## [1,] FALSE FALSE FALSE FALSE FALSE
## [2,] FALSE FALSE FALSE FALSE FALSE
## [3,] FALSE FALSE FALSE FALSE FALSE
## [4,] FALSE FALSE FALSE FALSE FALSE
## [5,] FALSE FALSE FALSE FALSE FALSE
## [6,] FALSE FALSE FALSE FALSE FALSE
## object_id has_VC has_angel has_roundA has_roundB has_roundC has_roundD
## [1,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [2,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [3,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [4,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [5,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [6,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## avg_participants is_top500 Gender Age Experience Education C1 C2
## [1,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [2,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [3,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [4,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [5,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [6,] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## C3 C4 status
## [1,] FALSE FALSE FALSE
## [2,] FALSE FALSE FALSE
## [3,] FALSE FALSE FALSE
## [4,] FALSE FALSE FALSE
## [5,] FALSE FALSE FALSE
## [6,] FALSE FALSE FALSE## ID state_code latitude
## 0.00000 0.00000 0.00000
## longitude zip_code id
## 0.00000 0.00000 0.00000
## city Unnamed..6 name
## 0.00000 0.00000 0.00000
## labels founded_at closed_at
## 0.00000 0.00000 0.00000
## first_funding_at last_funding_at age_first_funding_year
## 0.00000 0.00000 0.00000
## age_last_funding_year age_first_milestone_year age_last_milestone_year
## 0.00000 16.46804 16.46804
## relationships funding_rounds funding_total_usd
## 0.00000 0.00000 0.00000
## milestones state_code.1 is_CA
## 0.00000 0.00000 0.00000
## is_NY is_MA is_TX
## 0.00000 0.00000 0.00000
## is_otherstate category_code is_software
## 0.00000 0.00000 0.00000
## is_web is_mobile is_enterprise
## 0.00000 0.00000 0.00000
## is_advertising is_gamesvideo is_ecommerce
## 0.00000 0.00000 0.00000
## is_biotech is_consulting is_othercategory
## 0.00000 0.00000 0.00000
## object_id has_VC has_angel
## 0.00000 0.00000 0.00000
## has_roundA has_roundB has_roundC
## 0.00000 0.00000 0.00000
## has_roundD avg_participants is_top500
## 0.00000 0.00000 0.00000
## Gender Age Experience
## 0.00000 0.00000 0.00000
## Education C1 C2
## 0.00000 0.00000 0.00000
## C3 C4 status
## 0.00000 0.00000 0.00000#the dataset df, has lots of cells with missing data entries
#viewing actual missing data
md.pattern(df)
## ID state_code latitude longitude zip_code id city Unnamed..6 name labels
## 771 1 1 1 1 1 1 1 1 1 1
## 152 1 1 1 1 1 1 1 1 1 1
## 0 0 0 0 0 0 0 0 0 0
## founded_at closed_at first_funding_at last_funding_at
## 771 1 1 1 1
## 152 1 1 1 1
## 0 0 0 0
## age_first_funding_year age_last_funding_year relationships funding_rounds
## 771 1 1 1 1
## 152 1 1 1 1
## 0 0 0 0
## funding_total_usd milestones state_code.1 is_CA is_NY is_MA is_TX
## 771 1 1 1 1 1 1 1
## 152 1 1 1 1 1 1 1
## 0 0 0 0 0 0 0
## is_otherstate category_code is_software is_web is_mobile is_enterprise
## 771 1 1 1 1 1 1
## 152 1 1 1 1 1 1
## 0 0 0 0 0 0
## is_advertising is_gamesvideo is_ecommerce is_biotech is_consulting
## 771 1 1 1 1 1
## 152 1 1 1 1 1
## 0 0 0 0 0
## is_othercategory object_id has_VC has_angel has_roundA has_roundB
## 771 1 1 1 1 1 1
## 152 1 1 1 1 1 1
## 0 0 0 0 0 0
## has_roundC has_roundD avg_participants is_top500 Gender Age Experience
## 771 1 1 1 1 1 1 1
## 152 1 1 1 1 1 1 1
## 0 0 0 0 0 0 0
## Education C1 C2 C3 C4 status age_first_milestone_year
## 771 1 1 1 1 1 1 1
## 152 1 1 1 1 1 1 0
## 0 0 0 0 0 0 152
## age_last_milestone_year
## 771 1 0
## 152 0 2
## 152 304Looking at the two attributes that do have actual missing value entries, we have it that the two are having a total missing values of 16.46804% in total. The imputation of the two data attributes is as the code lines that follow below;
#imputing data
df$age_first_milestone_year[which(is.na(df$age_first_milestone_year))]=
mean(df$age_first_milestone_year,na.rm = TRUE)
df$age_last_milestone_year[which(is.na(df$age_last_milestone_year))]=
mean(df$age_last_milestone_year,na.rm = TRUE)
#farther cleaning requires that irrelevant attributes are dropped
df=df[, -c(3,4,5,8,23,40)]
head(df)## ID state_code id city name labels founded_at
## 1 1 CA c:6669 San Diego Bandsintown 1 1/1/2007
## 2 2 CA c:16283 Los Gatos TriCipher 1 1/1/2000
## 3 3 CA c:65620 San Diego Plixi 1 3/18/2009
## 4 4 CA c:42668 Cupertino Solidcore Systems 1 1/1/2002
## 5 5 CA c:65806 San Francisco Inhale Digital 0 8/1/2010
## 6 6 CA c:22898 Mountain View Matisse Networks 0 1/1/2002
## closed_at first_funding_at last_funding_at age_first_funding_year
## 1 4/1/2009 1/1/2010 2.2493
## 2 2/14/2005 12/28/2009 5.1260
## 3 3/30/2010 3/30/2010 1.0329
## 4 2/17/2005 4/25/2007 3.1315
## 5 10/1/2012 8/1/2010 4/1/2012 0.0000
## 6 2/15/2009 7/18/2006 7/18/2006 4.5452
## age_last_funding_year age_first_milestone_year age_last_milestone_year
## 1 3.0027 4.6685 6.7041
## 2 9.9973 7.0055 7.0055
## 3 1.0329 1.4575 2.2055
## 4 5.3151 6.0027 6.0027
## 5 1.6685 0.0384 0.0384
## 6 4.5452 5.0027 5.0027
## relationships funding_rounds funding_total_usd milestones is_CA is_NY is_MA
## 1 3 3 375000 3 1 0 0
## 2 9 4 40100000 1 1 0 0
## 3 5 1 2600000 2 1 0 0
## 4 5 3 40000000 1 1 0 0
## 5 2 2 1300000 1 1 0 0
## 6 3 1 7500000 1 1 0 0
## is_TX is_otherstate category_code is_software is_web is_mobile
## 1 0 0 music 0 0 0
## 2 0 0 enterprise 0 0 0
## 3 0 0 web 0 1 0
## 4 0 0 software 1 0 0
## 5 0 0 games_video 0 0 0
## 6 0 0 network_hosting 0 0 0
## is_enterprise is_advertising is_gamesvideo is_ecommerce is_biotech
## 1 0 0 0 0 0
## 2 1 0 0 0 0
## 3 0 0 0 0 0
## 4 0 0 0 0 0
## 5 0 0 1 0 0
## 6 0 0 0 0 0
## is_consulting is_othercategory has_VC has_angel has_roundA has_roundB
## 1 0 1 0 1 0 0
## 2 0 0 1 0 0 1
## 3 0 0 0 0 1 0
## 4 0 0 0 0 0 1
## 5 0 0 1 1 0 0
## 6 0 1 0 0 0 1
## has_roundC has_roundD avg_participants is_top500 Gender Age
## 1 0 0 1.0000 0
## 2 1 1 4.7500 1 Prefer not to say 41-50
## 3 0 0 4.0000 1 Male 31-40
## 4 1 1 3.3333 1 Male 31-40
## 5 0 0 1.0000 1 Male 31-40
## 6 0 0 3.0000 1 Male 31-40
## Experience Education C1 C2 C3 C4 status
## 1 6-8 Years PhD Yes Yes Yes Yes acquired
## 2 Over 8 Years Masters Yes Yes Yes Yes acquired
## 3 Over 8 Years Bachelors Yes Yes Yes Yes acquired
## 4 8-10 Years Masters Yes Yes Yes Yes acquired
## 5 8-10 Years Masters Yes No Yes No closed
## 6 Over 8 Years Bachelors Yes Yes Yes Yes closedThe next step that would follow would be the Descriptive Research Design and for this matter therefore, there will be a run of the descriptive analytics and to start off it is wise to note that the dataset has a mix of integer, numeric and character attributes in the mix up.
## 'data.frame': 923 obs. of 51 variables:
## $ ID : int 1 2 3 4 5 6 7 8 9 10 ...
## $ state_code : Factor w/ 35 levels "AR","AZ","CA",..: 3 3 3 3 3 3 3 3 13 3 ...
## $ id : Factor w/ 922 levels "c:10054","c:101312",..: 812 171 808 593 809 311 162 699 15 675 ...
## $ city : Factor w/ 221 levels "Acton","Addison",..: 173 109 173 56 174 126 126 178 215 144 ...
## $ name : Factor w/ 922 levels "#waywire","1000memories",..: 77 815 613 743 367 469 665 147 859 217 ...
## $ labels : int 1 1 1 1 0 0 1 1 1 1 ...
## $ founded_at : Factor w/ 217 levels "1/1/1984","1/1/1985",..: 16 9 107 11 184 11 14 13 11 151 ...
## $ closed_at : Factor w/ 203 levels "","1/1/2001",..: 1 1 1 1 26 75 1 1 1 1 ...
## $ first_funding_at : Factor w/ 585 levels "1/1/2000","1/1/2001",..: 296 217 279 219 491 460 559 514 507 206 ...
## $ last_funding_at : Factor w/ 680 levels "1/1/2001","1/1/2004",..: 8 206 323 380 341 534 298 112 281 274 ...
## $ age_first_funding_year : num 2.25 5.13 1.03 3.13 0 ...
## $ age_last_funding_year : num 3 10 1.03 5.32 1.67 ...
## $ age_first_milestone_year: num 4.6685 7.0055 1.4575 6.0027 0.0384 ...
## $ age_last_milestone_year : num 6.7041 7.0055 2.2055 6.0027 0.0384 ...
## $ relationships : int 3 9 5 5 2 3 6 25 13 14 ...
## $ funding_rounds : int 3 4 1 3 2 1 3 3 3 3 ...
## $ funding_total_usd : num 375000 40100000 2600000 40000000 1300000 7500000 26000000 34100000 9650000 5750000 ...
## $ milestones : int 3 1 2 1 1 1 2 3 4 4 ...
## $ is_CA : int 1 1 1 1 1 1 1 1 0 1 ...
## $ is_NY : int 0 0 0 0 0 0 0 0 0 0 ...
## $ is_MA : int 0 0 0 0 0 0 0 0 1 0 ...
## $ is_TX : int 0 0 0 0 0 0 0 0 0 0 ...
## $ is_otherstate : int 0 0 0 0 0 0 0 0 0 0 ...
## $ category_code : Factor w/ 35 levels "advertising",..: 20 9 35 31 12 21 31 11 19 35 ...
## $ is_software : int 0 0 0 1 0 0 1 0 0 0 ...
## $ is_web : int 0 0 1 0 0 0 0 0 0 1 ...
## $ is_mobile : int 0 0 0 0 0 0 0 0 1 0 ...
## $ is_enterprise : int 0 1 0 0 0 0 0 0 0 0 ...
## $ is_advertising : int 0 0 0 0 0 0 0 0 0 0 ...
## $ is_gamesvideo : int 0 0 0 0 1 0 0 0 0 0 ...
## $ is_ecommerce : int 0 0 0 0 0 0 0 0 0 0 ...
## $ is_biotech : int 0 0 0 0 0 0 0 0 0 0 ...
## $ is_consulting : int 0 0 0 0 0 0 0 0 0 0 ...
## $ is_othercategory : int 1 0 0 0 0 1 0 1 0 0 ...
## $ has_VC : int 0 1 0 0 1 0 1 0 1 1 ...
## $ has_angel : int 1 0 0 0 1 0 0 0 0 1 ...
## $ has_roundA : int 0 0 1 0 0 0 1 1 1 1 ...
## $ has_roundB : int 0 1 0 1 0 1 1 1 0 0 ...
## $ has_roundC : int 0 1 0 1 0 0 0 0 0 0 ...
## $ has_roundD : int 0 1 0 1 0 0 0 1 1 0 ...
## $ avg_participants : num 1 4.75 4 3.33 1 ...
## $ is_top500 : int 0 1 1 1 1 1 1 1 1 1 ...
## $ Gender : Factor w/ 4 levels "","Female","Male",..: 1 4 3 3 3 3 3 3 3 3 ...
## $ Age : Factor w/ 54 levels "","31-40","41-100",..: 1 5 2 2 2 2 2 2 2 2 ...
## $ Experience : Factor w/ 3 levels "6-8 Years","8-10 Years",..: 1 3 3 2 2 3 3 3 3 3 ...
## $ Education : Factor w/ 3 levels "Bachelors","Masters",..: 3 2 1 2 2 1 1 1 1 2 ...
## $ C1 : Factor w/ 1 level "Yes": 1 1 1 1 1 1 1 1 1 1 ...
## $ C2 : Factor w/ 2 levels "No","Yes": 2 2 2 2 1 2 2 2 2 2 ...
## $ C3 : Factor w/ 2 levels "No","Yes": 2 2 2 2 2 2 1 2 2 1 ...
## $ C4 : Factor w/ 2 levels "No","Yes": 2 2 2 2 1 2 1 2 2 1 ...
## $ status : Factor w/ 2 levels "acquired","closed": 1 1 1 1 2 2 1 1 1 1 ...## ID state_code id city
## Min. : 1.0 CA :488 c:28482 : 2 San Francisco:128
## 1st Qu.:231.5 NY :106 c:10054 : 1 New York : 91
## Median :462.0 MA : 83 c:101312: 1 Mountain View: 47
## Mean :462.0 TX : 42 c:10137 : 1 Palo Alto : 35
## 3rd Qu.:692.5 WA : 42 c:10153 : 1 Austin : 27
## Max. :923.0 CO : 19 c:10158 : 1 Santa Clara : 27
## (Other):143 (Other) :916 (Other) :568
## name labels founded_at closed_at
## Redwood Systems: 2 Min. :0.0000 1/1/2003: 55 :588
## #waywire : 1 1st Qu.:0.0000 1/1/2002: 54 6/1/2013: 25
## 1000memories : 1 Median :1.0000 1/1/2005: 54 1/1/2012: 24
## 41st Parameter : 1 Mean :0.6468 1/1/2006: 54 7/1/2013: 15
## 5min Media : 1 3rd Qu.:1.0000 1/1/2000: 53 5/1/2013: 12
## Aardvark : 1 Max. :1.0000 1/1/2004: 50 1/1/2011: 8
## (Other) :916 (Other) :603 (Other) :251
## first_funding_at last_funding_at age_first_funding_year
## 1/1/2008 : 18 1/1/2008 : 11 Min. :-9.0466
## 1/1/2007 : 16 1/1/2012 : 6 1st Qu.: 0.5767
## 1/1/2006 : 12 12/1/2007: 6 Median : 1.4466
## 3/1/2006 : 10 9/1/2008 : 6 Mean : 2.2356
## 12/1/2006: 9 4/1/2012 : 5 3rd Qu.: 3.5753
## 2/1/2005 : 9 7/1/2008 : 5 Max. :21.8959
## (Other) :849 (Other) :884
## age_last_funding_year age_first_milestone_year age_last_milestone_year
## Min. :-9.047 Min. :-14.170 Min. :-7.005
## 1st Qu.: 1.670 1st Qu.: 1.252 1st Qu.: 2.930
## Median : 3.529 Median : 3.055 Median : 4.754
## Mean : 3.931 Mean : 3.055 Mean : 4.754
## 3rd Qu.: 5.560 3rd Qu.: 4.003 3rd Qu.: 6.040
## Max. :21.896 Max. : 24.685 Max. :24.685
##
## relationships funding_rounds funding_total_usd milestones
## Min. : 0.000 Min. : 1.000 Min. :1.100e+04 Min. :0.000
## 1st Qu.: 3.000 1st Qu.: 1.000 1st Qu.:2.725e+06 1st Qu.:1.000
## Median : 5.000 Median : 2.000 Median :1.000e+07 Median :2.000
## Mean : 7.711 Mean : 2.311 Mean :2.542e+07 Mean :1.842
## 3rd Qu.:10.000 3rd Qu.: 3.000 3rd Qu.:2.472e+07 3rd Qu.:3.000
## Max. :63.000 Max. :10.000 Max. :5.700e+09 Max. :8.000
##
## is_CA is_NY is_MA is_TX
## Min. :0.0000 Min. :0.0000 Min. :0.00000 Min. :0.0000
## 1st Qu.:0.0000 1st Qu.:0.0000 1st Qu.:0.00000 1st Qu.:0.0000
## Median :1.0000 Median :0.0000 Median :0.00000 Median :0.0000
## Mean :0.5276 Mean :0.1148 Mean :0.08992 Mean :0.0455
## 3rd Qu.:1.0000 3rd Qu.:0.0000 3rd Qu.:0.00000 3rd Qu.:0.0000
## Max. :1.0000 Max. :1.0000 Max. :1.00000 Max. :1.0000
##
## is_otherstate category_code is_software is_web
## Min. :0.000 software :153 Min. :0.0000 Min. :0.000
## 1st Qu.:0.000 web :144 1st Qu.:0.0000 1st Qu.:0.000
## Median :0.000 mobile : 79 Median :0.0000 Median :0.000
## Mean :0.221 enterprise : 73 Mean :0.1658 Mean :0.156
## 3rd Qu.:0.000 advertising: 62 3rd Qu.:0.0000 3rd Qu.:0.000
## Max. :1.000 games_video: 52 Max. :1.0000 Max. :1.000
## (Other) :360
## is_mobile is_enterprise is_advertising is_gamesvideo
## Min. :0.00000 Min. :0.00000 Min. :0.00000 Min. :0.00000
## 1st Qu.:0.00000 1st Qu.:0.00000 1st Qu.:0.00000 1st Qu.:0.00000
## Median :0.00000 Median :0.00000 Median :0.00000 Median :0.00000
## Mean :0.08559 Mean :0.07909 Mean :0.06717 Mean :0.05634
## 3rd Qu.:0.00000 3rd Qu.:0.00000 3rd Qu.:0.00000 3rd Qu.:0.00000
## Max. :1.00000 Max. :1.00000 Max. :1.00000 Max. :1.00000
##
## is_ecommerce is_biotech is_consulting is_othercategory
## Min. :0.00000 Min. :0.00000 Min. :0.00000 Min. :0.0000
## 1st Qu.:0.00000 1st Qu.:0.00000 1st Qu.:0.00000 1st Qu.:0.0000
## Median :0.00000 Median :0.00000 Median :0.00000 Median :0.0000
## Mean :0.02709 Mean :0.03684 Mean :0.00325 Mean :0.3229
## 3rd Qu.:0.00000 3rd Qu.:0.00000 3rd Qu.:0.00000 3rd Qu.:1.0000
## Max. :1.00000 Max. :1.00000 Max. :1.00000 Max. :1.0000
##
## has_VC has_angel has_roundA has_roundB
## Min. :0.0000 Min. :0.0000 Min. :0.0000 Min. :0.0000
## 1st Qu.:0.0000 1st Qu.:0.0000 1st Qu.:0.0000 1st Qu.:0.0000
## Median :0.0000 Median :0.0000 Median :1.0000 Median :0.0000
## Mean :0.3261 Mean :0.2546 Mean :0.5081 Mean :0.3922
## 3rd Qu.:1.0000 3rd Qu.:1.0000 3rd Qu.:1.0000 3rd Qu.:1.0000
## Max. :1.0000 Max. :1.0000 Max. :1.0000 Max. :1.0000
##
## has_roundC has_roundD avg_participants is_top500
## Min. :0.0000 Min. :0.00000 Min. : 1.000 Min. :0.0000
## 1st Qu.:0.0000 1st Qu.:0.00000 1st Qu.: 1.500 1st Qu.:1.0000
## Median :0.0000 Median :0.00000 Median : 2.500 Median :1.0000
## Mean :0.2329 Mean :0.09967 Mean : 2.839 Mean :0.8093
## 3rd Qu.:0.0000 3rd Qu.:0.00000 3rd Qu.: 3.800 3rd Qu.:1.0000
## Max. :1.0000 Max. :1.00000 Max. :16.000 Max. :1.0000
##
## Gender Age Experience Education
## : 52 31-40 :819 6-8 Years : 52 Bachelors:562
## Female : 51 : 52 8-10 Years :104 Masters :309
## Male :768 41-100 : 1 Over 8 Years:767 PhD : 52
## Prefer not to say: 52 41-101 : 1
## 41-50 : 1
## 41-51 : 1
## (Other): 48
## C1 C2 C3 C4 status
## Yes:923 No :256 No :153 No :307 acquired:597
## Yes:667 Yes:770 Yes:616 closed :326
##
##
##
##
## ## [1] "Those that were acquired totaled to 597 and those that were closed totaled to 326. From this data summary and summation, it is as clear, that a considerable number of firms had struggled in their operations and were eventually closed whereas others must have been promising enough but due to lack of enough funds of operations were instead acquired. The main reasons for acquisition will be largely expounded on under the discussion section. To this point though, it is only wise to note that the results do dictate the fact that, there is a clear view that more firms were performing close to better and looked promising and due to this fact, could easily be selected for acquisition"## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 923 2.24 2.51 1.45 1.89 1.77 -9.05 21.9 30.94 2.1 9.97 0.08## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 923 3.93 2.97 3.53 3.65 2.88 -9.05 21.9 30.94 1.09 3.12 0.1## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 923 3.06 2.72 3.06 2.83 2.19 -14.17 24.68 38.85 1.03 7.27 0.09## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 923 4.75 2.94 4.75 4.6 2.58 -7.01 24.68 31.69 0.78 3.04 0.1## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 923 7.71 7.27 5 6.5 4.45 0 63 63 2.32 8.56 0.24## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 923 2.31 1.39 2 2.1 1.48 1 10 9 1.35 2.23 0.05## vars n mean sd median trimmed mad min max
## X1 1 923 25419749 189634364 1e+07 13381063 12883794 11000 5.7e+09
## range skew kurtosis se
## X1 5699989000 29.06 865.75 6241891## vars n mean sd median trimmed mad min max range skew kurtosis se
## X1 1 923 1.84 1.32 2 1.75 1.48 0 8 8 0.58 0.25 0.04#Getting to know the statistical distribution through Histograms
The dataset from the above histogram plot, clearly indicates a normally distributed data points with the peak at the center and yet the shorter data points spread out to the ends of the data points
For the above two cases, most firms, got their first and last funding in the periods that were between 0 and 5 years on total. These years are the most crucial years of firm and a firm in most cases does not look at expanding its operations, but looks forward into expanding its operations instead. The firms that receive funding years later than 5 years of existence on the other side tends to be seeking expansion funds and that is allowed as this stage needs greater investments.
The number of relationships as indicated by the above graph has it that of the firms that portrayed levels of having relationships with other firms, most firms in the list had relationship periods that spanned from 0 years to 5 years and that number kept dropping and from here the data is highly skewed to the right.
The data for the number of funding given to firms indicates that, number if round that firms were funded ranged between 1 and 2 and then the data frequency drops and therefore the dataset is skewed to the right. Most firms are just well on their own after the first or even second funding and that’s why there are fewer funding frequencies as numbers move along.
Doing bi-variate analysis
#doing bi-variate analysis
#we will do on continuous and categorical
#we will do on categorical and categorical
#we will do on continuous and continous
?by
by(df$funding_total_usd, df$status, summary)## df$status: acquired
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 5.970e+04 4.575e+06 1.270e+07 3.105e+07 2.706e+07 5.700e+09
## ------------------------------------------------------------
## df$status: closed
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 11000 1000000 5000000 15115322 16525000 510000000## df$status: acquired
## [1] 31046622
## ------------------------------------------------------------
## df$status: closed
## [1] 15115322## df$status: acquired
## [1] 12700000
## ------------------------------------------------------------
## df$status: closed
## [1] 5e+06## df$status: acquired
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.000 1.000 2.000 2.162 3.000 8.000
## ------------------------------------------------------------
## df$status: closed
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.000 0.000 1.000 1.255 2.000 6.000## df$status: acquired
## [1] 2.162479
## ------------------------------------------------------------
## df$status: closed
## [1] 1.254601## df$status: acquired
## [1] 2
## ------------------------------------------------------------
## df$status: closed
## [1] 1
#Allocating Equipment
#Doing probability plots
# Filter for firm types
dr1=df %>%
filter(status=='acquired')
qqnorm(dr1$milestones)
qqline(dr1$milestones, col=3)
dr1=df %>%
filter(status=='acquired')
qqnorm(dr1$funding_total_usd)
qqline(dr1$funding_total_usd, col=3)
Density plot
#density plots: with these plots the intensity of an entry in a column is noticeable by it's overiding nature
df %>%
filter(status=='closed' | status=='acquired') %>%
ggplot(aes(x=funding_total_usd, fill = status)) +
geom_density(alpha=0.9, color='darkblue') +
ggtitle('funding_total_usd') 
df %>%
filter(C2=='Yes' | status=='No') %>%
ggplot(aes(x=funding_total_usd, fill = status)) +
geom_density(alpha=0.9, color='darkblue') +
ggtitle('C2') 
df %>%
filter(C4=='Yes' | status=='No') %>%
ggplot(aes(x=funding_total_usd, fill = status)) +
geom_density(alpha=0.9, color='darkblue') +
ggtitle('C4')
Building Linear Regression Model
#linear regression is to asses the statistically significant attributes for building classifiction algorithms
mod1=lm(milestones~age_first_funding_year+
age_last_funding_year+
age_first_milestone_year+
age_last_milestone_year+
relationships+
funding_rounds+
funding_total_usd, data=df)
summary(mod1)#gives information on the figures from linear regression including error rates##
## Call:
## lm(formula = milestones ~ age_first_funding_year + age_last_funding_year +
## age_first_milestone_year + age_last_milestone_year + relationships +
## funding_rounds + funding_total_usd, data = df)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.6120 -0.5293 0.0425 0.5834 3.6984
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 8.947e-01 8.668e-02 10.322 < 2e-16 ***
## age_first_funding_year -4.403e-02 3.090e-02 -1.425 0.15457
## age_last_funding_year -7.552e-02 2.913e-02 -2.592 0.00968 **
## age_first_milestone_year -2.353e-01 2.038e-02 -11.543 < 2e-16 ***
## age_last_milestone_year 2.804e-01 1.850e-02 15.156 < 2e-16 ***
## relationships 7.149e-02 5.226e-03 13.681 < 2e-16 ***
## funding_rounds 7.958e-02 3.892e-02 2.045 0.04117 *
## funding_total_usd -2.657e-10 1.732e-10 -1.534 0.12533
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9848 on 915 degrees of freedom
## Multiple R-squared: 0.4498, Adjusted R-squared: 0.4456
## F-statistic: 106.9 on 7 and 915 DF, p-value: < 2.2e-16##status Column will be taken as our independent variable.
#Splitting will be done with respect to status
split_model=sample.split(df,SplitRatio = 0.70)
train=subset(df,split_model==TRUE)
test=subset(df,split_model==FALSE)
nrow(train)## [1] 634## [1] 289#now lets proceed and develop logistic regression
modlog=glm(formula = status~age_first_funding_year+
age_last_funding_year+
age_first_milestone_year+
age_last_milestone_year+
relationships+
funding_rounds+
funding_total_usd+milestones+
avg_participants,
data = train,family = 'binomial')
#predicting the values using the test dataset
resultlog=predict(modlog,newdata = test,type = 'response')
head(resultlog)## 5 8 13 19 23 24
## 0.69664595 0.01081804 0.78656464 0.12708446 0.78663663 0.27325968## [1] 0.001978452 1.000000000##from the range of probabilities above we can now get the confusion matrix
#for us to do that we develop a table of the actual and the predicted values
table(test$status,resultlog>0.5)##
## FALSE TRUE
## acquired 158 32
## closed 42 57##
## FALSE TRUE
## acquired 135 55
## closed 29 70It is clearly evident that, only the attributes age_first_funding_year and funding_total_usd are not statistically significant and for that matter and the remaining attributes, those that are statistically significant will be used for building the classification algorithms.
#predicting the values using the test dataset
resultlog=predict(modlog, newdata = test, type = 'response')
head(resultlog)## 5 8 13 19 23 24
## 0.69664595 0.01081804 0.78656464 0.12708446 0.78663663 0.27325968log_new=cbind(Actual=test$status,Predicted=resultlog)
#finding accuracy and errors
log_new=as.data.frame(log_new)
#getting the error in prediction(subtract the predicted values from the actual values)
error1=log_new$Actual-log_new$Predicted
head(error1)## [1] 1.3033540 0.9891820 1.2134354 0.8729155 1.2133634 1.7267403## Actual Predicted error1
## 5 2 0.69664595 1.3033540
## 8 1 0.01081804 0.9891820
## 13 2 0.78656464 1.2134354
## 19 1 0.12708446 0.8729155
## 23 2 0.78663663 1.2133634
## 24 2 0.27325968 1.7267403## [1] 1.07117rmse1=sqrt(mean((log_new$error1)^2))
#Mean Absolute Error
library(MLmetrics)
MAE=MAE(log_new$Actual,log_new$Predicted)
MAE## [1] 0.9856348## [1] 0.3569257## [1] 3.264751## NULLUsing the statistically significant attributes and splitting the main dataset in a 70:30 ratio for train and test sets respectively, the developed model was trained using 632 rows and tested on a test data with 291 rows. Of the developed model, the accuracy rate was ranging between values of 6.234853e-06 and 8.097985e-01.
library(ROCR)
#Developing an ROCR curve as a measure of accuracy
PL=prediction(resultlog,test$status)
ACC=performance(PL,"acc")
plot(ACC)
From the above codeline, the highest value of perfromance is when the threshhold is at 0.58 With the accuracy degree of 0.74 which is 74%. For there to be highest level and most accurate predictions.
##Building a Decision Tree
library(tree)
tree.df=tree(status~age_first_funding_year+
age_last_funding_year+
age_first_milestone_year+
age_last_milestone_year+
relationships+
funding_rounds+
funding_total_usd+milestones+
avg_participants,
data=df)
summary(tree.df)##
## Classification tree:
## tree(formula = status ~ age_first_funding_year + age_last_funding_year +
## age_first_milestone_year + age_last_milestone_year + relationships +
## funding_rounds + funding_total_usd + milestones + avg_participants,
## data = df)
## Variables actually used in tree construction:
## [1] "relationships" "funding_total_usd"
## [3] "milestones" "age_last_milestone_year"
## [5] "age_last_funding_year" "age_first_funding_year"
## [7] "avg_participants"
## Number of terminal nodes: 10
## Residual mean deviance: 0.9438 = 861.7 / 913
## Misclassification error rate: 0.2243 = 207 / 923
##Using Split Data
tree.df1=tree(status~age_first_funding_year+
age_last_funding_year+
age_first_milestone_year+
age_last_milestone_year+
relationships+
funding_rounds+
funding_total_usd+milestones+
avg_participants,
data=train)
plot(tree.df1)
text(tree.df1,pretty = 0)
##Prediciting values
tree.pred=predict(tree.df1, test, type = "class")
#confusion matrix
table(test$status,tree.pred)## tree.pred
## acquired closed
## acquired 183 7
## closed 61 38## $size
## [1] 9 8 5 2 1
##
## $dev
## [1] 163 160 159 164 227
##
## $k
## [1] -Inf 0.000000 1.333333 7.000000 69.000000
##
## $method
## [1] "misclass"
##
## attr(,"class")
## [1] "prune" "tree.sequence"
#predicting values again
tree.pred=predict(tree.df1, test, type = "class")
table(test$status,tree.pred)## tree.pred
## acquired closed
## acquired 183 7
## closed 61 38##Building Random Forest
library(randomForest)
set.seed(111)
rf1=randomForest(status~age_first_funding_year+
age_last_funding_year+
age_first_milestone_year+
age_last_milestone_year+
relationships+
funding_rounds+
funding_total_usd+milestones+
avg_participants,
data=train)
rf1##
## Call:
## randomForest(formula = status ~ age_first_funding_year + age_last_funding_year + age_first_milestone_year + age_last_milestone_year + relationships + funding_rounds + funding_total_usd + milestones + avg_participants, data = train)
## Type of random forest: classification
## Number of trees: 500
## No. of variables tried at each split: 3
##
## OOB estimate of error rate: 21.92%
## Confusion matrix:
## acquired closed class.error
## acquired 368 39 0.0958231
## closed 100 127 0.4405286## rp1
## acquired closed
## acquired 174 16
## closed 34 65
## MeanDecreaseGini
## age_first_funding_year 36.20969
## age_last_funding_year 31.33394
## age_first_milestone_year 27.17028
## age_last_milestone_year 38.22115
## relationships 56.50446
## funding_rounds 14.78873
## funding_total_usd 42.08283
## milestones 23.11928
## avg_participants 20.88417## [1] 7201 6978 5702 6029 5665 3206 7633 2831 4810##K-Nearst Neighbor
# Libraries
library(caret)
library(pROC)
library(mlbench)
# KNN Model
trControl <- trainControl(method = "repeatedcv",
number = 10,
repeats = 3,
classProbs = TRUE,
summaryFunction = twoClassSummary)
set.seed(222)
fit <- train(status ~ age_first_funding_year+
age_last_funding_year+
age_first_milestone_year+
age_last_milestone_year+
relationships+
funding_rounds+
funding_total_usd+milestones+
avg_participants,
data = train,
method = 'knn',
tuneLength = 20,
trControl = trControl,
preProc = c("center", "scale"),
metric = "ROC",
tuneGrid = expand.grid(k = 1:60))
# Model Performance
fit## k-Nearest Neighbors
##
## 634 samples
## 9 predictor
## 2 classes: 'acquired', 'closed'
##
## Pre-processing: centered (9), scaled (9)
## Resampling: Cross-Validated (10 fold, repeated 3 times)
## Summary of sample sizes: 571, 571, 572, 570, 570, 571, ...
## Resampling results across tuning parameters:
##
## k ROC Sens Spec
## 1 0.6662039 0.7951220 0.5372859
## 2 0.7027739 0.7936789 0.5390646
## 3 0.7225433 0.8444309 0.5155468
## 4 0.7326698 0.8329268 0.5364295
## 5 0.7437672 0.8575813 0.5125823
## 6 0.7554646 0.8592073 0.5099473
## 7 0.7628925 0.8648780 0.5258893
## 8 0.7674427 0.8673374 0.5389987
## 9 0.7700594 0.8738211 0.5170619
## 10 0.7720584 0.8722358 0.5272069
## 11 0.7734950 0.8796138 0.5215415
## 12 0.7712766 0.8665854 0.5169302
## 13 0.7742161 0.8788008 0.5198287
## 14 0.7770114 0.8713211 0.5125823
## 15 0.7786278 0.8780081 0.5185771
## 16 0.7774859 0.8731098 0.5214756
## 17 0.7770979 0.8821341 0.5095520
## 18 0.7789911 0.8804472 0.5065217
## 19 0.7771332 0.8771545 0.5094862
## 20 0.7791949 0.8804065 0.5080369
## 21 0.7803944 0.8755285 0.5021739
## 22 0.7811409 0.8714228 0.4991436
## 23 0.7839361 0.8763415 0.5050725
## 24 0.7839984 0.8754878 0.4991436
## 25 0.7833236 0.8771545 0.5021080
## 26 0.7844370 0.8853455 0.5065876
## 27 0.7854299 0.8820732 0.5050725
## 28 0.7877190 0.8836585 0.5094862
## 29 0.7895708 0.8869715 0.5065876
## 30 0.7902137 0.8927236 0.5080369
## 31 0.7907341 0.8943293 0.5109354
## 32 0.7909993 0.8984146 0.5094203
## 33 0.7928511 0.8984146 0.5095520
## 34 0.7942122 0.8976220 0.5050725
## 35 0.7950387 0.9025813 0.5110672
## 36 0.7948743 0.9026016 0.5020422
## 37 0.7964964 0.9034553 0.5065876
## 38 0.7972950 0.8992886 0.5066535
## 39 0.7990403 0.9010163 0.5066535
## 40 0.7993030 0.9034756 0.5007246
## 41 0.8002782 0.9010163 0.5063900
## 42 0.7994470 0.9026220 0.5036232
## 43 0.7977072 0.9017683 0.5050066
## 44 0.7989044 0.9009146 0.5065217
## 45 0.7993054 0.9001220 0.5065876
## 46 0.7975124 0.8976016 0.5110672
## 47 0.7966311 0.8967886 0.5024374
## 48 0.7955276 0.8968699 0.5007246
## 49 0.7957459 0.8984959 0.5052701
## 50 0.7965302 0.8976829 0.5067194
## 51 0.7970009 0.8968699 0.5022398
## 52 0.7966643 0.8952439 0.5022398
## 53 0.7955690 0.8968699 0.5038867
## 54 0.7959459 0.8976626 0.4965086
## 55 0.7963677 0.8976829 0.4935441
## 56 0.7971509 0.8935976 0.4936100
## 57 0.7979647 0.8960366 0.4949934
## 58 0.7985234 0.8960366 0.4965086
## 59 0.7990948 0.8927846 0.4921607
## 60 0.7995663 0.8927642 0.4877470
##
## ROC was used to select the optimal model using the largest value.
## The final value used for the model was k = 41.
## ROC curve variable importance
##
## Importance
## relationships 100.00
## milestones 71.57
## funding_rounds 50.77
## age_last_milestone_year 49.35
## funding_total_usd 40.54
## avg_participants 37.29
## age_first_milestone_year 28.30
## age_last_funding_year 17.16
## age_first_funding_year 0.00## Confusion Matrix and Statistics
##
## Reference
## Prediction acquired closed
## acquired 172 45
## closed 18 54
##
## Accuracy : 0.782
## 95% CI : (0.7299, 0.8282)
## No Information Rate : 0.6574
## P-Value [Acc > NIR] : 2.573e-06
##
## Kappa : 0.4822
##
## Mcnemar's Test P-Value : 0.001054
##
## Sensitivity : 0.9053
## Specificity : 0.5455
## Pos Pred Value : 0.7926
## Neg Pred Value : 0.7500
## Prevalence : 0.6574
## Detection Rate : 0.5952
## Detection Prevalence : 0.7509
## Balanced Accuracy : 0.7254
##
## 'Positive' Class : acquired
## ###Niave Bayes Classification
# Libraries
library(naivebayes)
library(dplyr)
library(ggplot2)
library(psych)
# Naive Bayes Model
md=naive_bayes(status ~ age_first_funding_year+
age_last_funding_year+
age_first_milestone_year+
age_last_milestone_year+
relationships+
funding_rounds+
funding_total_usd+milestones+
avg_participants, data = train, usekernel = T)
md##
## ================================== Naive Bayes ==================================
##
## Call:
## naive_bayes.formula(formula = status ~ age_first_funding_year +
## age_last_funding_year + age_first_milestone_year + age_last_milestone_year +
## relationships + funding_rounds + funding_total_usd + milestones +
## avg_participants, data = train, usekernel = T)
##
## ---------------------------------------------------------------------------------
##
## Laplace smoothing: 0
##
## ---------------------------------------------------------------------------------
##
## A priori probabilities:
##
## acquired closed
## 0.6419558 0.3580442
##
## ---------------------------------------------------------------------------------
##
## Tables:
##
## ---------------------------------------------------------------------------------
## ::: age_first_funding_year::acquired (KDE)
## ---------------------------------------------------------------------------------
##
## Call:
## density.default(x = x, na.rm = TRUE)
##
## Data: x (407 obs.); Bandwidth 'bw' = 0.5093
##
## x y
## Min. :-2.0264 Min. :6.751e-05
## 1st Qu.: 0.7464 1st Qu.:2.152e-02
## Median : 3.5192 Median :6.654e-02
## Mean : 3.5192 Mean :9.007e-02
## 3rd Qu.: 6.2920 3rd Qu.:1.309e-01
## Max. : 9.0648 Max. :2.905e-01
##
## ---------------------------------------------------------------------------------
## ::: age_first_funding_year::closed (KDE)
## ---------------------------------------------------------------------------------
##
## Call:
## density.default(x = x, na.rm = TRUE)
##
## Data: x (227 obs.); Bandwidth 'bw' = 0.7533
##
## x y
## Min. :-11.306 Min. :8.000e-08
## 1st Qu.: -2.713 1st Qu.:4.083e-04
## Median : 5.879 Median :2.328e-03
## Mean : 5.879 Mean :2.906e-02
## 3rd Qu.: 14.472 3rd Qu.:2.013e-02
## Max. : 23.065 Max. :2.224e-01
##
## ---------------------------------------------------------------------------------
## ::: age_last_funding_year::acquired (KDE)
## ---------------------------------------------------------------------------------
##
## Call:
## density.default(x = x, na.rm = TRUE)
##
## Data: x (407 obs.); Bandwidth 'bw' = 0.6932
##
## x y
## Min. :-2.493 Min. :1.663e-05
## 1st Qu.: 1.626 1st Qu.:7.015e-03
## Median : 5.745 Median :4.243e-02
## Mean : 5.745 Mean :6.063e-02
## 3rd Qu.: 9.864 3rd Qu.:1.167e-01
## Max. :13.984 Max. :1.511e-01
##
## ---------------------------------------------------------------------------------
## ::: age_last_funding_year::closed (KDE)
## ---------------------------------------------------------------------------------
##
## Call:
## density.default(x = x, na.rm = TRUE)
##
## Data: x (227 obs.); Bandwidth 'bw' = 0.9765
##
## x y
## Min. :-11.976 Min. :5.430e-06
## 1st Qu.: -3.048 1st Qu.:8.298e-04
## Median : 5.879 Median :2.983e-03
## Mean : 5.879 Mean :2.797e-02
## 3rd Qu.: 14.807 3rd Qu.:3.354e-02
## Max. : 23.735 Max. :1.462e-01
##
## ---------------------------------------------------------------------------------
## ::: age_first_milestone_year::acquired (KDE)
## ---------------------------------------------------------------------------------
##
## Call:
## density.default(x = x, na.rm = TRUE)
##
## Data: x (407 obs.); Bandwidth 'bw' = 0.6144
##
## x y
## Min. :-5.5939 Min. :1.811e-05
## 1st Qu.:-0.4819 1st Qu.:3.391e-03
## Median : 4.6302 Median :1.506e-02
## Mean : 4.6302 Mean :4.886e-02
## 3rd Qu.: 9.7422 3rd Qu.:7.780e-02
## Max. :14.8542 Max. :2.016e-01
##
## ---------------------------------------------------------------------------------
## ::: age_first_milestone_year::closed (KDE)
## ---------------------------------------------------------------------------------
##
## Call:
## density.default(x = x, na.rm = TRUE)
##
## Data: x (227 obs.); Bandwidth 'bw' = 0.5613
##
## x y
## Min. :-8.689 Min. :1.000e-08
## 1st Qu.:-1.930 1st Qu.:6.107e-04
## Median : 4.829 Median :4.143e-03
## Mean : 4.829 Mean :3.695e-02
## 3rd Qu.:11.588 3rd Qu.:3.847e-02
## Max. :18.347 Max. :2.791e-01
##
## ---------------------------------------------------------------------------------
## ::: age_last_milestone_year::acquired (KDE)
## ---------------------------------------------------------------------------------
##
## Call:
## density.default(x = x, na.rm = TRUE)
##
## Data: x (407 obs.); Bandwidth 'bw' = 0.7206
##
## x y
## Min. :-5.9126 Min. :1.571e-05
## 1st Qu.:-0.4974 1st Qu.:2.453e-03
## Median : 4.9178 Median :2.446e-02
## Mean : 4.9178 Mean :4.612e-02
## 3rd Qu.:10.3330 3rd Qu.:8.403e-02
## Max. :15.7482 Max. :1.638e-01
##
## ---------------------------------------------------------------------------------
## ::: age_last_milestone_year::closed (KDE)
## ---------------------------------------------------------------------------------
##
## Call:
## density.default(x = x, na.rm = TRUE)
##
## Data: x (227 obs.); Bandwidth 'bw' = 0.7199
##
## x y
## Min. :-9.165 Min. :2.781e-05
## 1st Qu.:-1.753 1st Qu.:1.268e-03
## Median : 5.659 Median :2.571e-03
## Mean : 5.659 Mean :3.370e-02
## 3rd Qu.:13.071 3rd Qu.:4.343e-02
## Max. :20.483 Max. :2.384e-01
##
## ---------------------------------------------------------------------------------
## ::: relationships::acquired (KDE)
## ---------------------------------------------------------------------------------
##
## Call:
## density.default(x = x, na.rm = TRUE)
##
## Data: x (407 obs.); Bandwidth 'bw' = 1.817
##
## x y
## Min. :-5.452 Min. :4.770e-06
## 1st Qu.:13.024 1st Qu.:4.150e-04
## Median :31.500 Median :1.555e-03
## Mean :31.500 Mean :1.352e-02
## 3rd Qu.:49.976 3rd Qu.:1.503e-02
## Max. :68.452 Max. :8.707e-02
##
## ---------------------------------------------------------------------------------
## ::: relationships::closed (KDE)
## ---------------------------------------------------------------------------------
##
## Call:
## density.default(x = x, na.rm = TRUE)
##
## Data: x (227 obs.); Bandwidth 'bw' = 0.7943
##
## x y
## Min. :-2.383 Min. :6.220e-06
## 1st Qu.: 5.559 1st Qu.:9.871e-04
## Median :13.500 Median :7.235e-03
## Mean :13.500 Mean :3.145e-02
## 3rd Qu.:21.441 3rd Qu.:4.189e-02
## Max. :29.383 Max. :1.569e-01
##
## ---------------------------------------------------------------------------------
##
## # ... and 4 more tables
##
## ---------------------------------------------------------------------------------
## acquired closed ID state_code id city
## 1 0.8994251 0.10057493 1 CA c:6669 San Diego
## 2 0.9765062 0.02349377 2 CA c:16283 Los Gatos
## 3 0.5495379 0.45046212 3 CA c:65620 San Diego
## 4 0.9346075 0.06539251 4 CA c:42668 Cupertino
## 6 0.3733746 0.62662545 6 CA c:22898 Mountain View
## 7 0.9589714 0.04102858 7 CA c:16191 Mountain View
## name labels founded_at closed_at first_funding_at
## 1 Bandsintown 1 1/1/2007 4/1/2009
## 2 TriCipher 1 1/1/2000 2/14/2005
## 3 Plixi 1 3/18/2009 3/30/2010
## 4 Solidcore Systems 1 1/1/2002 2/17/2005
## 6 Matisse Networks 0 1/1/2002 2/15/2009 7/18/2006
## 7 RingCube Technologies 1 1/1/2005 9/21/2006
## last_funding_at age_first_funding_year age_last_funding_year
## 1 1/1/2010 2.2493 3.0027
## 2 12/28/2009 5.1260 9.9973
## 3 3/30/2010 1.0329 1.0329
## 4 4/25/2007 3.1315 5.3151
## 6 7/18/2006 4.5452 4.5452
## 7 3/18/2010 1.7205 5.2110
## age_first_milestone_year age_last_milestone_year relationships funding_rounds
## 1 4.6685 6.7041 3 3
## 2 7.0055 7.0055 9 4
## 3 1.4575 2.2055 5 1
## 4 6.0027 6.0027 5 3
## 6 5.0027 5.0027 3 1
## 7 3.0000 6.6082 6 3
## funding_total_usd milestones is_CA is_NY is_MA is_TX is_otherstate
## 1 375000 3 1 0 0 0 0
## 2 40100000 1 1 0 0 0 0
## 3 2600000 2 1 0 0 0 0
## 4 40000000 1 1 0 0 0 0
## 6 7500000 1 1 0 0 0 0
## 7 26000000 2 1 0 0 0 0
## category_code is_software is_web is_mobile is_enterprise is_advertising
## 1 music 0 0 0 0 0
## 2 enterprise 0 0 0 1 0
## 3 web 0 1 0 0 0
## 4 software 1 0 0 0 0
## 6 network_hosting 0 0 0 0 0
## 7 software 1 0 0 0 0
## is_gamesvideo is_ecommerce is_biotech is_consulting is_othercategory has_VC
## 1 0 0 0 0 1 0
## 2 0 0 0 0 0 1
## 3 0 0 0 0 0 0
## 4 0 0 0 0 0 0
## 6 0 0 0 0 1 0
## 7 0 0 0 0 0 1
## has_angel has_roundA has_roundB has_roundC has_roundD avg_participants
## 1 1 0 0 0 0 1.0000
## 2 0 0 1 1 1 4.7500
## 3 0 1 0 0 0 4.0000
## 4 0 0 1 1 1 3.3333
## 6 0 0 1 0 0 3.0000
## 7 0 1 1 0 0 1.6667
## is_top500 Gender Age Experience Education C1 C2 C3 C4
## 1 0 6-8 Years PhD Yes Yes Yes Yes
## 2 1 Prefer not to say 41-50 Over 8 Years Masters Yes Yes Yes Yes
## 3 1 Male 31-40 Over 8 Years Bachelors Yes Yes Yes Yes
## 4 1 Male 31-40 8-10 Years Masters Yes Yes Yes Yes
## 6 1 Male 31-40 Over 8 Years Bachelors Yes Yes Yes Yes
## 7 1 Male 31-40 Over 8 Years Bachelors Yes Yes No No
## status
## 1 acquired
## 2 acquired
## 3 acquired
## 4 acquired
## 6 closed
## 7 acquired#confusion matrix
#Confusion Matrix - train data
p1 <- predict(md, train)
(tab1 <- table(p1, train$status))##
## p1 acquired closed
## acquired 324 60
## closed 83 167## [1] 0.2255521Conclusion-
Random Forest Model gives most accurate prediction at 76% for a company’s success or failure
There are more firms that were acquired than those that were closed
Most Managers Seem to be between age 31 to 40
A positive association exists between Project Success and Entrepreneurial Acclimatization
A positive association exists between Acclimatization in Technology and Success in Projects
A positive arbitration that exists in the relationship of Entrepreneurial Acclimatization and Project Success