Hi,
Central to my PhD is the construction of a credit-rating model for toll facilities (roads, bridges and tunnels).
I am about to review 60 Rating Agency credit reports (between 10 and 25 pages each) to extract the key quantitative and qualitative rating 'drivers'. Before I start this (considerable!) task I would appreciate any general thoughts list members have on an appropriate approach.
My idea is simply to work through the reports and, for each asset, note the attributes that contribute (positively and negatively) to the rating agency's determination of credit strength. As well as identifying these attributes, I will - where possible - 'size' them (eg. attribute = debt service coverage ratio; size = a minimum of 1.3x: or attribute = tariff escalation formula; size = CPI - x: etc.). I intend to build an Access database as I go along in which each record (line) will be an individual toll facility and the fields will develop as I identify 'new' attributes. This would seem to fit with my ultimate objective of constructing my rating model (to be specified later by the data).
This seems a logical approach, but there must be other incidences when folks have had to work through reports (containing quantitative and qualitative information) and 'extract' data or build a data-set? I just wanted to check - before embarking on my task - to see if there was a better (or, indeed, 'best practice') approach? I don't want to reinvent the wheel or, worse, realise after report number 37 that there was a better way of doing things!
Any and all thoughts gratefully received - perhaps off-list to save the bandwidth.
Many thanks in advance.
Robert Bain
ITS, Leeds
|