I don't think you'll be able to simulate clicking on a webpage with matlab, and even if you could, it's totally the wrong approach.
Ideally, there would be a way to query the database directly and get the output as xml, json, or some other format easily parsable by a computer. This is the first thing I'd look for if I were you.
Otherwise, a better approach is to simply generate the correct URL for each query. For example, the result of your first search is always the page: https://www.sec.gov/cgi-bin/series?&sc=companyseries&ticker= TICKERVALUE. So it's trivial to obtain its content with:
baseurl = 'https://www.sec.gov/cgi-bin/series';
tickervalue = 'CSEFX'; %for example
cikpage = webread(baseurl, 'sc', 'companyseries', 'ticker', tickervalue);
Unfortunately, you get the output as html, which is not easy to parse for information. As I said, it would be better to be able to get the result as xml, or json, or ... Otherwise, the following regex may work (no guarantee it will work on every page):
cikvalue = regexp(cikpage, '(?<=<td colspan="3" nowrap="nowrap"><a class="search" href="[^"]*">)\d+', 'match', 'once')
To get to the next page, it's then simply:
baseurl = 'https://www.sec.gov/cgi-bin/browse-edgar';
searchresult = webread(baseurl, 'CIK', cikvalue, 'action', 'getcompany');
You then have to parse the resulting html page to get the next link. Something you'll have to figure out for yourself...
