Three mixes, one correlation matrix and a shortfall reported openly: the GF570 Unit 4 asset allocation model for a composite foundation ends 0.94 points short in compound terms. Searches like "gf 570 unit 4 assignment example", "gf570 unit 4 sample" and "gf570 unit 4 example" land here.
What a finished GF570 Unit 4 asset allocation model looks like
A model of about six pages with three exhibits: capital market assumptions for seven asset classes, a correlation matrix and a candidate comparison. Expected returns run from 3.5 percent for cash to 11.0 for private equity, volatilities from 1.0 to 24.0. Mix A, with 60 percent in growth assets, expects 7.16 percent at 9.97 percent volatility. Mix B, at 73 percent, expects 7.71 at 11.87. Mix C, at 85 percent, expects 8.30 at 13.84, above the ceiling. Converted to compound terms by subtracting half the variance, B yields 7.01 and C 7.35, so the arithmetic figure that seemed to satisfy the policy for C does not survive. A risk budget shows equities and private equity in B supplying 88.5 percent of risk from 65 percent of capital.
How a GF570 Unit 4 example is structured
Assumptions come first, each labeled as the adviser's forward estimate rather than history, with a sentence on where such estimates usually come from. The correlation matrix follows in full, and a paragraph picks out the three relationships that matter most: equities with private equity, core bonds with inflation-linked bonds, and US equity with core bonds. Each candidate is then computed in the same order: weights, expected return, variance from the matrix, volatility, compound return and a fifth-percentile one-year outcome. A comparison table sets all three beside the policy's tests. The correlation argument is made numerically, weighted-average volatility against portfolio volatility, 14.34 against 11.87 for B. Risk contributions by class follow. The recommendation adopts B, states the 0.94 shortfall and invokes the policy clause sending that gap to the board.
Assumptions labeled as estimates
Every expected return and volatility is marked as a forward view, and a note explains why starting yield predicts bond returns better than past returns do.
Three relationships that matter
Correlations of 0.75 between public and private equity, 0.65 between core and inflation-linked bonds, and 0.10 between US equity and core bonds carry most of the result.
Arithmetic against compound
Subtracting half the variance turns C's 8.30 into 7.35 and B's 7.71 into 7.01, which is the comparison a compound objective actually requires.
Diversification measured
B's volatility of 11.87 sits 2.47 points below the weighted average of its parts, the figure that shows correlation doing the work.
Where the risk lives
US equity supplies 40.6 percent of B's risk, non-US equity 31.0 and private equity 16.9, while core and inflation-linked bonds together add barely 3.
A shortfall sent upward
Mix B misses the objective by 0.94 points in compound terms, and the model passes that gap to the board as the policy's clause requires.
Where marks go in GF570 Unit 4
Allocation models lose most when diversification is claimed from the number of asset classes rather than shown from correlations, since the unit asks for the reasoning explicitly. Comparing an arithmetic expected return with a compound objective is the next most costly error, and it is easy to miss because the arithmetic figure flatters. Models that choose the mix meeting the return target while quietly breaching the risk ceiling misread the policy they were built to serve. Assumptions presented as history, or left without any source, draw comment. A risk budget is not always required, but papers showing where risk actually sits tend to outscore those reporting capital weights alone. Hiding a shortfall, rather than reporting it as the policy directs, costs more than the shortfall itself.
Get a GF570 Unit 4 example written to your instructions
Capital market assumptions, the policy limits and the rubric from your GF570 Unit 4 prompt are what the model runs on; if your instructor supplies a correlation matrix, it is used unchanged. Candidates are tested against every limit, and compound return is compared with the objective. First custom sample at no cost, usually within 24-48h.
GF570 Unit 4 questions, answered
Where do capital market assumptions come from?
Usually from the course materials, an instructor's table or published forward-looking estimates by large asset managers. The sample labels its figures as composite forward estimates and explains the logic behind two of them. If your prompt supplies assumptions, the custom model uses them unchanged, since graders compare your results with the answer built from their own inputs.
Why compare compound rather than arithmetic return with the objective?
Because the objective describes growth over many years, and wealth compounds. Volatility drags compound growth below the arithmetic average by roughly half the variance. In the sample that drag turns an apparently sufficient 8.30 percent into 7.35. Comparing the wrong figure is a frequent reason an allocation seems to meet a policy it does not.
Do I need an optimizer?
Rarely. Most GF570 prompts ask for a few candidates tested against the policy, and a spreadsheet handles the matrix arithmetic. The sample tests three mixes and shows the variance calculation for one in full. Where your prompt asks for an efficient frontier, the custom model adds one and explains why the chosen mix sits where it does on that curve.