All Collections
Feature Deep Dives
FAQ: Optimization and Portfolio Lab
FAQ: Optimization and Portfolio Lab
K
Written by Kyle Reinhardt
Updated over a week ago

Rev. Date June 5, 2023

Please note, this feature is only available in certain jurisdictions. Additional fees may apply.

## What is optimization?

Optimization is a mathematical analysis that is commonly run by allocators to help with exercises such as position sizing and portfolio rebalancing. At a high-level, optimization seeks to find the combination of investments that best meets an allocator’s defined investment objective.

## How can Venn help with optimization analyses?

Venn permits allocators to run time-consuming mathematical optimization analyses quickly and easily, using Venn’s simple and intuitive interface. Using Venn’s optimizer, you can:

• Estimate investment or strategy allocations using the forecasts your organization set on Venn

• Easily modify your organization’s investment objective and allocation and factor constraints to instantly generate new optimization results

• Compare optimization results against your organization’s portfolios and benchmarks

## What are some of the basics of optimization?

The scope of the optimization is set to the portfolio you have specified. You can select: (1) “maximize returns” with a volatility less than x%; (2) “target returns” of y% with minimum volatility; or (3) “maximize Sharpe” with a return greater than z%. Upon selecting the optimization objective, the default level of volatility (or return) shown is the forecasted volatility (or forecasted return) of the portfolio.t will be given the maximum allocation corresponding to the level in the portfolio above it.

## Generally, how does Venn approach optimization?

Venn takes a factor-based approach to optimization. Using the Two Sigma Factor Lens, Venn decomposes investment return streams into a set of factor exposures and a residual. Venn then runs optimization computations using your organization’s specified objectives and constraints, considering the contribution to risk and return for each investment based on your forecasts.

For information on how to customize forecasts, refer to this help article.

## What type of optimization does Venn use?

Venn uses a variant of a mean-variance optimization that utilizes a factor risk model. We believe a well-specified and appropriately constrained mean-variance optimization can generate intuitive results.

## What investments can be included in optimization?

You can include investments available for analysis on Venn in optimization, so long as they contain at least 1 year of returns.[1] Venn does not include any investments in an optimization analysis that were not explicitly added by you.

Additionally, within “Portfolio Lab,” you have the option to add a set of new opportunities to optimization. This “bench” of investments you’ve selected will be considered for new allocations in an optimized portfolio should Venn believe any of them can help achieve your desired objective and constraints.

## What is Portfolio Lab?

Portfolio Lab is a space that allows you to explore multiple simultaneous portfolios in the optimization process. While a single optimized solution is shown, alternate portfolios are also visible. You can analyze certain tradeoffs of each portfolio by clicking on them within the “Efficient Frontier.” Once you select an alternate portfolio, the page will refresh and you can see how it compares to your current portfolio. You can also evaluate a “bench” of new opportunities within the context of an optimized portfolio.

## What allocation constraints can be set?

By default, all investments and strategies within the portfolio are unconstrained, so the optimization results may not be realistic. For example, you may have certain allocations in your portfolio that are not liquid and are not desirable to trade out of. Alternatively, you might have a target level of factor exposure that the portfolio needs to be optimized around (e.g. you want to ensure the portfolio has at least 0.5 beta to the Equity factor).

At any level of the portfolio, you can choose to set minimum and maximum constraints by \$ or % as well as by factor exposure. The lock button allows you to keep the allocation of an investment or strategy unchanged from its current allocation.

Constraints can be set by clicking “Allocation” or “Exposure” icons near the upper left hand side of Portfolio Lab. You can choose to set constraints around both the size of allocations and factor exposures, at all levels of the portfolio (investment, sub-strategies, strategies, etc.). As default, Venn will indicate the level of factor exposures of the current portfolio as reference.

The constraints are accessible within “Portfolio Lab” and within your Portfolio Policy under “Manage Data.”

If you are running multiple optimizations on the same portfolio with a set of baseline constraints that must be in place, it could also be useful to set a Portfolio Policy. You can set your Portfolio Policy, which is a set of allocation and factor exposure constraints, by clicking on Manage Data (gear icon), clicking “Portfolio Policy” and adding the constraints as needed. This setting will always apply to the specific portfolio without having to enter the constraints repeatedly each time a new optimization analysis is run.

When making any changes to the constraints on top of the set Portfolio Policy, the inputs panel will indicate if the current set of constraints are still in agreement with the Portfolio Policy.

## What can I do if the optimizer is unable to find a solution?

It is easy to overly constrain the optimizer such that it can’t find a solution. You can try loosening any of your constraints to see if that leads to possible results. Examples include:

• Increase the volatility cap (e.g. from 3.0% to 5.0%, etc.)

• Lower the target return (e.g. from 5.0% to 4.5%, etc.)

• Widen the factor constraints (e.g. from 0 - .20 on Equity to 0 - .30, etc.)

• Unlock investments or strategies

• Increase the max allocation for investments or strategies

## Can I optimize excess returns or Tracking Error?

Currently, Portfolio Lab is only able to target return, volatility, or Sharpe Ratio for optimization and does not support benchmark-relative optimization.

## Should I trade based on optimization results?

When your organization makes investment decisions, Venn’s optimization results may be used only as a resource supplementing your organization’s primary investment, tax, accounting, legal and other advisors and your organization’s due diligence. Optimization does not take into account your organization’s overall or complete financial situation, level of financial sophistication, investment experience, investment mandate, financial goals, or investment profile. Venn’s optimizer does not accept all relevant investment profile or other data, make all potential settings available or customizable, or otherwise provide all important or relevant analysis to an investment decision, nor does Venn or Two Sigma Investor Solutions inquire into or otherwise determine if any data input, settings selected or analyses run are actually appropriate for your organization. No determination has been made regarding the suitability of any securities, other financial instruments, portfolios, allocations, strategies, objectives, constraints, etc. for any organization.

In addition, Venn does not account for any fees or expenses, such as transaction, financing, management or performance fees, associated with making portfolio changes or holding any investments, or for tax, accounting, legal, regulatory or other considerations. Please see Venn’s Subscriber Agreement and Two Sigma Investor Solution’s Form ADV (publicly available and available on Venn) for further information on limitations of Venn.

​[1] The minimum amount of data required depends on data frequency and type of analysis. For Optimization, if a user’s portfolio or investment has daily or monthly data, Venn requires 1 year at a minimum but will default to periods with more data based on your forecast settings. For quarterly data, Venn requires 3 years at a minimum (or 12 data points) but will default to periods with more data based on your forecast settings to run analysis. Portfolios or investments with quarterly data must be interpolated to perform Optimization.

This document highlights certain aspects of this feature. As an overview, it does not discuss all material facts or assumptions. Please see Important Disclosure and Disclaimer Information.