Data Integration Strategy

Do take the necessary time and attention to the details before you start your projects. There are different strategies with regards to the data integration. Please note that the primary imperative with any of those strategies is that the positions tally with the banks positions, i.e the positions must match. Secondly, please note that we get 2 main sets of data, namely Position Data and Transaction Data.

Strategy 1: Primary data source is Transaction and Position data is only used for Position Reconciliation --> Transaction Based

In this case, we use only transaction data to import the movements/actions. This is the most tedious work as there are unlimited types of transactions which all have to be mapped. For example, we will have to have all existing transaction types mapped, such as non-performance related deposits and withdrawals mapped among many other types so that the cash positions etc tally the banks. Most of our clients do NOT have the patience or resources to work with us analysing and reviewing the data and process. Therefore this strategy is very seldom used.

Strategy 2: Primary data source is the Position File and no Transaction data is used at all --> Position Increment

This is the simplest strategy, as we are using only position files. In this strategy, the positions which we receive for a date is compared to the positions in the system and the differential is posted as action. For example, if today the custodian says the cash position is 1,000 USD, but our system shows 1,200 USD, we post -200 USD (inflow/outflow, or sometimes called position change) to our system. Mind you, the 200 USD difference may not be due to a single movement, but a result of many movements. This strategy has the advantage that we will get the positions right (with the high certainty) and can be implemented very very fast. The disadvantage is that we don't have the granularity and details. One of the important downsides is that we are not able to capture the Transfers (non-performance related deposits/withdrawals).

Strategy 3: Primary data source is the Position File and use Transaction data to 'quantize' --> Quantization

In order to combine the benefit of the 2 worlds, we have a strategy called 'quantization'. In this strategy, we start by using the Position Increment strategy, so that positions are reflected correctly and with minimal lead time. Thereafter, depending on how responsive our clients are in analyzing and reviewing with us, we incrementally 'quantize' the position with transactions. For example, say the client is interested in getting the correct performance, what we would do is to look at the transaction file and only focus on the transaction types which are deemed "transfers" without having to completely analyze and understand the transaction file together with the client. We only pick transaction types which are transfers, send our analysis to our client for verification and then do the following in the mapping process:

Before start start 'quantizing' we know that the cash position is correct and we do not want to compromise the integrity of the positions. Therefore, when we create 2 additional entries (+ Transfer, - Amendment) of the amount of which the cash transaction is deemed Transfer by the custodian. Our system normalized the NAV's by Transfers which it finds in the portfolio.