[R] How to manipulate tables

[David Winsemius]

On Dec 26, 2009, at 2:02 PM, James Rome wrote:

> I am sorry to be bothering the list so much.
>
> I made a table of counts of flight arrivals by hour:

No, you made a list of tables, which is different.

> cnts=tapply(Arrival4,list(Hour),table). There are up to 15 arrivals  
> in a
> bin.

Why not work with something more regular, like (this is all guesswork  
and untested in absence of test data objects):

table(Arrival4, factor(Hour, levels=1:15)  )    ?
# or use whatever your max(arrivals) might be.

 > xx <- factor(sample(1:10, 4))
 > xx
[1] 9 4 7 1
Levels: 1 4 7 9

 > table(factor(xx, levels=1:10) )

  1  2  3  4  5  6  7  8  9 10
  1  0  0  1  0  0  1  0  1  0



cnts
>
> $`0`
>
> 1  2  3  4  5  6  7  8  9 10 13
> 1  2  5  9  2  7  5  4  2  4  1
>
> $`1`
>
> 1 2 3 4
> 3 2 2 1
>
> $`2`
>
> 1 3
> 2 2
> .  . .
>
> My first problem is how to get this table filled in with the 0 slots.
> E.g., I want
> $`0`
>
> 1  2  3  4  5  6  7  8  9 10 11 12 13 14 15
> 1  2  5  9  2  7  5  4  2  4   0   0   1   0   0
>
> for all 24 hours. The elements of the tables are lists, but I do not
> seem to be able to extract the names of each list, which I think would
> allow this manipulation.
>
> My second problem is that I want to compute probability  
> distributions. I
> have
> lambda
>          0           1           2           4           5
> 6           7
> 0.199190283 0.013765182 0.006477733 0.017813765 0.093117409  
> 0.160323887
> 0.401619433
>          8           9          10          11          12
> 13          14
> 0.191093117 0.177327935 0.318218623 0.404858300 0.463157895  
> 0.495546559
> 0.435627530
>         15          16          17          18          19
> 20          21
> 0.418623482 0.307692308 0.405668016 0.484210526 0.580566802  
> 0.585425101
> 0.519028340
>         22          23
> 0.556275304 0.503643725
>
> and I need to calculate lambda**cnts for each bin, and each hour. I am
> also unsure of how to do this.

Can you give a justification for this procedure? (I'm not saying it is  
nonsense, only that it does not make make sense to me. So perhaps  
there is some mathematical property about which I need education.)

-- 

David Winsemius, MD
Heritage Laboratories
West Hartford, CT