Create tables - simple table construction in R, probably using stargazer package

I need in total 4 tables constructed. The tasks should be very easy for someone who knows R and how to create tables. Pictures are added for description on how I want them to look like.

Task 1: (short task)

Data = df1

Create a table that looks like picture 1, only also adding p-values to the very right in the table.

You need to calculate the mean, standard error and pvalue (H0 mean(variable A1)=0) of each of the variables [A1,B1,C,D1,E1]. Mean should be named = "CAR", standard error = "Std.err.", just like in the table, and pvalue ="P-value". Pvalue column should be added right of Std.err. Add stars (*) to denote any significance (if any). For the text above the table, just write: “Type text here”.



Variable name in df should be set to "Variable name in table"
A1 = [-20,-11]
B1 = [-10,-6]
C = [-5,10]
D1 = [11,20]
E1 = [21,60]

Task 2: (short task)

Data = df1

Now create a table that looks like picture 2.
ONLY variable C should be used when calculating mean (CAR) and standard error, Name this mean CAR[-5,10] just like in the picture. Add number of observations. No P-value column to add in this table, but add stars (*) to denote any significance. For the text above the table, just write: “Type text here”. Add title above text just like in the picture: “title”

Panel A:
Certified -“certified” variable ==1
noncertified -“certified” variable ==0

Panel B:
First-time issue  “first_time” variable ==1
Seasoned issue  “first_time” variable ==0

Panel C are not included! (just skip this)


Task 3:

Data = df2:

Create a table that looks like the one in picture 3:

All descriptions of variables should be equal to the ones in the picture. (NB! Environemntal materiality are not included here, just skip it).

In this table the greenbond==TRUE and nongreenbond==TRUE should be compared.

Calculate mean, (standard error), P-value for diff-in-means and number of observations and structure like in the picture.

Log(assets)= called “laggedsize” in df3
Return on assets = “laggedroa”
Tobin´s Q = “laggedtobinsq”
Leverage = “laggedleverage”

greenbond == TRUE Green bond issuer  used for column (1)
nongreenbond == TRUE  Non green bond issuer  used for column (2)

Then do the same for Environment rating (ASSET 4), Social rating (ASSET 4) and Governance rating (ASSET 4), but for only observations where the dummy variable “sufficient” == TRUE. That is: For green bond issuers  greenbond==TRUE & sufficient==TRUE, and for non-green bond issuers  nongreenbond==TRUE & sufficient==TRUE.

Environment rating (ASSET 4) = “laggedenv.score”
Social rating (ASSET 4) = “laggedsoc.score”
Governance rating (ASSET 4) = “laggedgov.score”


Task 4:

Data = df3

Create a table that looks like the one in picture 4:

In this table “greenbond” ==1 group should be compared against “match==1” group.

Items needed:
number of observations within each group
mean
median
std.dev
p-value for a difference in mean comparison between greenbond ==1 and match ==1, for each variable
p-value for a difference in median comparison between greenbond ==1 and match ==1, for each variable.

Variables to compare:
Log(assets)= called “laggedsize” in df3
Return on assets = “laggedroa”
Tobin´s Q = “laggedtobinsq”
Leverage = “laggedleverage”
∆ Log(assets) = “laggeddelta_size”
∆ Return on assets = “laggeddelta_roa”
∆ Tobin´s Q = “laggeddelta_tobinsq”
∆ Leverage = “laggeddelta_leverage”

For only when dummy “sufficient”==TRUE, compare greenbond==1 against match==1 group for these variables aswell.

Environment rating (ASSET 4) = “laggedenv.score”
Social rating (ASSET 4) = “laggedsoc.score”
Governance rating (ASSET 4) = “laggedgov.score”
∆ Environment rating (ASSET 4) = “laggeddelta_env.score”
∆ Social rating (ASSET 4) = “laggeddelta_soc.score”
∆ Governance rating (ASSET 4) = “laggeddelta_gov.score”
 
Variable to check:

Log(assets)= called “laggedsize” in df3
Return on assets = “laggedroa”
Leverage = “laggedleverage”