Both tools crunch large portfolios with hundreds of stock names in seconds to produce the optimal trading schedule of buys and sells. They do this by taking into account market impact costs, risk factor correlations and implied volatility as well as other data inputs.

On Monday, Instinet, the global agency broker, launched Wizard PRO (Portfolio Risk Optimizer), an implementation shortfall algorithm that takes into account sector correlations and individual stock correlations and natural hedges that exist between all the component stocks. “When you have a whole basket or portfolio of 100 or 200 stocks, they all individually have correlations,” explains Mike Plunkett, President of North America at Instinet in New York.

“What Wizard PRO does is it checks correlations between all equities and it will use those and then create a trading schedule around that that keeps your hedges in place,” says Plunkett. For example, if someone was buying GM and selling Ford, “It wouldn’t let you buy 100,000 GM if you couldn’t sell any Ford,” says Plunkett. He recommends using the tool (which is available through Instinet’s front ends Newport and Portal) for baskets of 15 or more stocks because it’s “difficult for any portfolio manager to trade that many stocks at one point in time and know how many correlations there are in the portfolio.”

On Tuesday, FlexTrade Systems introduced FlexPTS (portfolio trade scheduler), a trading solution that uses advanced optimization technology. “Definitely, there’s demand for it — It’s just evolution of technology,” says Vijay Kedia, CEO of FLEXTrade Systems, which operates a multi-asset execution management system (EMS) in Great Neck, NY.

“More and more firms have accepted that portfolio trading is the way to go,” says Kedia, especially for firms that are doing transition trading or executing larger portfolios, which take multiple days to complete, he adds. Firms such as FlexTrade are being forced to step ahead and are adding more features, he says.

The biggest benefit to FlexTrade’s tool is that it’s packaged in a broker-neutral system so that the buy side doesn’t have to disclose the whole portfolio to any broker, says Kedia.

“No buy-side would want to give the whole store away,” says Kedia. The same goes for other inputs that this requires, such as the buy-side firm’s short-term alpha expectations, which can be kept confidential, he adds. “You’re still doing the portfolio optimization, but you maintain the portfolio level schedule inside your own organization,” says FlexTrade’s CEO.

Interestingly, both the Instinet Wizard PRO and FlexTrade portfolio trading schedulers, utilize risk models from Northfield Information Services. Instinet relies on the well known U.S. Short -Term Equity Risk Model and Open Optimizer. “We wanted to take a best-of-breed approach,” says Thorsten Schmidt, First VP, algorithmic trading at Instinet. “Northfield does risk models for a living so we use their risk model data,” he says. Since Instinet has lots of transactions flowing through its system, it uses its own data on market impact to create a market impact model, adds Schmidt.

FlexTrade PTS employs its own cost model as well as the Northfield’s Short-Term Equity Risk Model, though Kedia says traders can plug in their own cost indicators because the system is flexibly architected to allow for that. “The two main inputs – the (market impact) cost model and the risk model could be replaced,” emphasizes Kedia, adding that different buy-side firms can use different models.

Once the trading schedule is created, the buy-side can execute the order anyway it likes, says FlexTrade’s CEO. “If there are 500 names, it can trade with 20 brokers, “he says. “You have full choice over which broker or DMA pipe.”

In Instinet’s case, the execution is automatic for the user, says Schmidt. “We just don’t send an order to one destination. We have these market interaction layers that take an order to be executed that split certain stocks for tactical reasons, he says. For an OTC stock, it may send a piece to an ECN or to Instinet’s crossing network or to CBX(Continuous Block Cross). “Once you change the characteristics you would re-optimize your portfolio and then it sends it out to the market again,” says Plunkett. Both portfolio-trading models allow the buy side to reoptimize their schedules on the fly.

“Once you change the characteristics you would reoptimize your portfolio, and then it sends it out to the market again,” says Plunkett. Both portfolio-trading models allow the buy side to reoptimize their schedules on the fly.