One of my course’s selective paper referee report writing assignment. I like this paper so, here’s the referee report I wrote (lighted edited) which comes with a crisp summarization that I find in times can be quite useful This is a beautifully written, intellectually clean paper. It uses traditional OR methods and is free of fancy deep learning. Bertsimas is a titan of the field. The paper is published in Management Science at an interesting timing of 2020. ...

I Am confused. More than baffled — Don’t know What — was going on — Ariana Emily – No Prisoner Be is a song cycle composed by Kevin Puts based on poems by Emily Dickinson. CSO’s 25/26 season resident artist Joyce DiDonato, who is also the OG artist thet premiered this work, performed the work Feb 12 on Chicago Symphony Center. The stage design and lighting is really fancy ...

I like this model. It’s cute. Unrealistic, but cute. Environment: Intermediate sector. A unit measure of firms $i \in [0,1]$. Firm $i$ produces differentiated variety $Y_i$ with linear technology: $$ Y_i = A_i L_i $$ We assume $A_i\equiv A$ for now. Final sector. A competitive firm aggregates varieties via CES: $$ Y = \left(\int_0^1 Y_i^{\frac{\sigma-1}{\sigma}}\, di\right)^{\frac{\sigma}{\sigma-1}}, \qquad \sigma > 1. $$ Consumer. Representative agent with preferences: $$ U = \frac{C^{1-\gamma}}{1-\gamma} - \frac{L^{1+\varphi}}{1+\varphi}. $$ Equilibrium Prices $(\{P_i\}_i, W, P)$ and allocations $(\{Y_i, \Pi_i, L_i\}_i, C, Y, L)$ such that: ...

Alchemist 101: don’t create gold. Define what you created as gold. Motivation: find model to predict economic facts Kaldor facts —capital-output ratio and factor shares are roughly constant over time. Kuznets facts — the sectoral composition of employment and consumption shifts systematically: Agriculture (A) down, Manufacturing (M) stable, while Services (S) goes up. Can a single neoclassical growth model deliver both sets of facts simultaneously? Yes, if we throw in the cauldron a non-homothetic preferences (Engel’s law). ...

All models are wrong, but some are useful. — George E. P. Box And, some model would have particularly peculiar assumptions that makes them easy and handy. Motivation The neoclassical growth model was designed to explain several facts observed in developed economies over long periods (the “Kaldor facts” from data from the U.S. (1900-2000)): Output per capita grows at roughly constant rate Capital-output, investment-output, and consumption-output ratios are roughly constant Interest rates are roughly constant Factor shares (capital vs labor income) are roughly constant Uzawa’s Theorem Consider $(K_t, L_t, C_t, Y_t)_t$ governed by: ...

Images for Orchestra L. 122 and La Mer. Chicago audience are warm, welcoming. We give a lot of standing ovations: Salonen conducts CSO for Debussy’s La mer. De ja vu: NY Phil opens Shanghai Summer Music Festival with French Impressionism Works See the post here Fun facts: Debussy, Turner, and Impressionism We also know that Debussy greatly admired Turner’s work [the English painter]. His richly atmospheric seascapes recorded the daily weather, the time of day, and even the most fleeting effects of wind and light in ways utterly new to painting, and they spoke directly to Debussy. (In 1902, when Debussy went to London, where he saw a number of Turner’s paintings, he enjoyed the trip but hated actually crossing the channel.) The name Debussy finally gave to the first section of La mer, From Dawn to Noon on the Sea, might easily be that of a painting by Turner made sixty years earlier, for the two shared not only a love of subject but also of long, specific, evocative titles. ...

Motivation Baumol (1967): Differential productivity growth across sectors drives “uneven growth” and structural transformation. As sectors experience different rates of technological progress, this causes relative prices to change and labor to reallocate. Question: Can this supply-side mechanism generate structural change while maintaining balanced growth (Kaldor facts)? Model Household utility: $$ \begin{align*} & \int_0^\infty \exp(-\rho t) \frac{c(t)^{1-\theta} - 1}{1-\theta} dt\cr &s.t. \ \dot K(t) = r(t)K(t) - w(t)\bar L -\sum_{i \in \lbrace A, M, S\rbrace}p^iC^i(t) \ \cr & \text{where, }c(t) = \left(\sum_{i \in \{A,S,M\}} \eta^i C^i(t)^{(\sigma-1)/\sigma}\right)^{\sigma/(\sigma-1)}. \end{align*} $$ Note that $\sigma$ represents elasticity of substitution (constant across sectors) ...

Motivation Paradox: Standard productivity-based theories (see Supply Side Structural Change in Macroeconomics) predict that sectors with higher productivity growth should have declining employment shares. This is counter-intuitive, and false in reality: Japan (1960-1990): Rapid manufacturing productivity growth → increased manufacturing GDP share A lot of empirical evidence (common knowledge…) shows, countries with faster manufacturing TFP growth did not experience faster manufacturing employment decline. And cross-country data shows no negative correlation between manufacturing productivity and manufacturing employment. [Source pending] ...

Late January and early Feburary is an intriguing time for birthdays… There’s Mozart (Jan 27 1756 - Dec 5 1791), Schubert (Jan 31 1797 - Nov 19 1828) and Felix Mendelssohn (Feb 3 1809 - Nov 4 1847) I had the fortune of playing the Midsummer Night’s Dream Overture when I was even too young to understand the beauty before internalizing it (16). But the spirit lives on. Happy birthday to one of my favorite composers! ...

The Hidden Weight of GMM Consistency Conditions Consider estimating parameter $\theta\in \Theta$ from data $\lbrace w_i\rbrace_{i \in [N]}$ Assume: Parameter space $\Theta\in \R^K$ is compact. The criterion function of GMM $$ > s_N(\theta) = s(\vec w; \theta) > $$ is continuous in $\theta$ $\forall, \vec w$. $s_N(\cdot)$ well behaves: $$ > \sup_{\theta\in \Theta}|s_N(\theta) - s_\infty(\theta) \xrightarrow{p}0. > $$ $s_\infty(\theta)$ has a unique minimum at $\theta_0$. Then $\hat \theta_{GMM} \xrightarrow{p}\theta$. The proof is essentially a topological argument — uniform convergence of continuous functions on a compact set, plus a unique minimum, pins down the limit of the minimizers. It is clean, elegant, and almost suspiciously general. ...