*Rev. Date August 2, 2022*

**What is the purpose of interpolation on Venn?**

Interpolation allows you to increase the return frequency from quarterly to daily for your private asset allocations. The two main benefits of interpolation include:

Private asset returns tend to be smoothed because of the illiquidity of the asset class, leading to artificially low volatility. Interpolating private asset data will provide you with what we believe to be a more accurate representation of the true risk of these types of allocations, by returning risk that is closer to that of the public market equivalent index.

The ability to analyze your total portfolio, inclusive of private assets, without losing the benefits of having more frequent data for investments that are not private assets. Previously, quarterly returns would have shifted all of the investments in your portfolio, regardless of their original frequency, to quarterly as well. This would then necessitate having 3+ years of full portfolio history to run factor-based analyses on Venn.[1]

**How can I interpolate my private asset returns on Venn?**

First, gather the historical quarterly returns for your private asset allocation by major category (according to the categories in the table below). Upload the returns to Venn. Once the upload is complete, open any analysis of that private asset allocation. In the top right of the analysis page, click “Interpolate” and select a private asset category or an investment of your choice to be used for interpolation.

**How does Venn interpolate my private asset allocations?**

After you upload your quarterly private asset return series and select its public market equivalent investment, here are the basic steps of interpolation:

Venn finds the public market equivalent index for the selected category.[2] See the table below for the mapping.

For each quarter, Venn will calculate a daily constant return that, when added to the public market equivalent index’s daily returns, will yield a quarterly return that closely matches the quarterly return you uploaded for that private asset.

The result is a daily return series with returns that closely match those of your original private asset return series and with volatility that roughly matches that of the public market equivalent index.

**What is the purpose of extrapolation on Venn?**

Extrapolation seeks to provide users with an estimate of the updated value of their private asset allocation. Because many private asset funds report their valuations on a meaningful lag, Venn will help fill the gap between the last reported valuation and today by estimating its valuation.

**How does Venn extrapolate my private asset allocations?**

After you upload your quarterly private asset return series and select its category, here are the basic steps of extrapolation:

Venn finds the public market equivalent index for the selected category. See the table above for the mapping.

Venn runs an Ordinary-Least-Squares regression of the private asset allocation’s quarterly returns on its public market equivalent index.[3]

Venn will then extrapolate from the last reported quarterly return to today using the in-sample beta coefficients from the regression multiplied by the out-of-sample returns of the public market equivalent index, plus the in-sample alpha.

**Where can I see extrapolation on Venn?**

The extrapolated returns for a private asset allocation can be viewed on the investment’s Cumulative Return chart under Performance Analysis.

**What are the limitations of interpolation and extrapolation on Venn?**

Any Venn user can analyze a single interpolated and/or extrapolated return series based on their uploaded returns but should keep in mind that the results, especially as it relates to risk measures like volatility and factor exposures, will depend heavily on the category and therefore the public market equivalent index selected.

For Venn Pro users only, interpolation and extrapolation can help quantify the likely risk of an overall private assets allocation (or subsets by category) to show how it fits within the rest of a multi-asset class portfolio.

In general, we believe these methods are best used when evaluating an entire private asset allocation in a specific category, such as Buyout or Venture Capital, and are most accurate when the allocation is “mature” (hence the data requirements below). This higher-level categorical analysis will tend to be less affected by the early negative returns of individual funds (i.e. J-curve effects) and will also have steadier value through time than individual funds.

**What are the minimum number of quarterly data points required for interpolation and extrapolation on Venn?**

Venn requires 3 years, or 12 quarterly returns, for interpolation and extrapolation.

**Why can’t I interpolate and extrapolate my private asset allocations?**

Assuming your private asset allocation has at least the minimum number of quarterly data points required, interpolation and extrapolation will be disabled in cases where the private asset series does not have a meaningfully and reliably positive relationship with any of the public market equivalent indices.

**Why are certain categories not recommended for interpolation and extrapolation?**

We will warn a user if interpolation and extrapolation are not recommended in cases where the private asset series does not have a meaningful and reliably positive relationship with a public market equivalent index.

In order for a category to be recommended, the private asset series must meet the following conditions in relation to the underlying public market equivalent:

beta must be >=.01

standard error of regression must be <=.4

[1] See How much data is required for factor analysis on Venn? for more information.

[2] Categories for the private asset categories were guided by research from Andrew Ang, Bingxu Chen, William N. Goetzmann, and Ludovic Phalippou (2014) in addition to research by Two Sigma.

[3] The extrapolation methodology also incorporates a lagged effect by using the sum of the most recent four quarters of the public market equivalent index as an additional independent variable in the regression.

*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.*