Tulitula commited on
Commit
ab0239b
Β·
verified Β·
1 Parent(s): 4e15a8a

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +54 -5
README.md CHANGED
@@ -26,7 +26,7 @@ LensIQ is a Gradio app for portfolio analysis using CAPM and the Capital Market
26
  - **Efficient mixes** (market/bills only)
27
  - **Same Οƒ** as your portfolio β†’ point on the CML at your Οƒ
28
  - **Same E[r]** as your portfolio β†’ Οƒ on the CML for that return
29
- - Weights shown so you can replicate the mixes.
30
 
31
  - **Suggestions**
32
  - Exactly **one** candidate per band (**Low / Medium / High**).
@@ -42,7 +42,56 @@ LensIQ is a Gradio app for portfolio analysis using CAPM and the Capital Market
42
 
43
  ## Run Locally
44
 
45
- ```bash
46
- docker build -t lensiq .
47
- docker run -p 7860:7860 -e GRADIO_SERVER_NAME=0.0.0.0 lensiq
48
- # Open http://localhost:7860
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
26
  - **Efficient mixes** (market/bills only)
27
  - **Same Οƒ** as your portfolio β†’ point on the CML at your Οƒ
28
  - **Same E[r]** as your portfolio β†’ Οƒ on the CML for that return
29
+ - Weights are shown so you can replicate the mixes.
30
 
31
  - **Suggestions**
32
  - Exactly **one** candidate per band (**Low / Medium / High**).
 
42
 
43
  ## Run Locally
44
 
45
+ docker build -t lensiq .
46
+ docker run -p 7860:7860 -e GRADIO_SERVER_NAME=0.0.0.0 lensiq
47
+ # Open http://localhost:7860
48
+
49
+ ---
50
+
51
+ ## Using the App
52
+
53
+ 1. Add tickers and include **VOO**.
54
+ 2. Enter **$ amounts** for each ticker.
55
+ 3. Set **Horizon** and **Lookback** (years).
56
+ 4. Click **Compute**.
57
+ 5. Review your portfolio, the two efficient mixes (with weights), and the **one** Low/Medium/High suggestion (with % and $).
58
+ 6. Download the generated CSV.
59
+
60
+ ---
61
+
62
+ ## Method
63
+
64
+ - Prices: Yahoo Finance monthly (`yfinance`)
65
+ - Risk-free: FRED tenor by horizon (1–30y)
66
+ - Market: `VOO`
67
+ - ERP: market CAPM E[r] βˆ’ rf
68
+ - Ξ²: covariance vs. market (excess returns)
69
+ - Οƒ (hist): annualized from monthly covariance
70
+ - Suggestions: 1,000 long-only mixes; embedding-aided ranking + MMR
71
+
72
+ ---
73
+
74
+ ## Synthetic Dataset
75
+
76
+ A CSV is written under `data/`, e.g. `data/investor_profiles_<timestamp>.csv`.
77
+
78
+ **Columns:** `tickers`, `weights`, `beta`, `mu_capm`, `sigma_hist`, `sigma_capm`
79
+
80
+ If you publish the dataset to Hugging Face, set the repo to:
81
+ `your-username/LensIQ-Synthetic-Portfolio-1000` and include this README.
82
+
83
+ ---
84
+
85
+ ## Project Structure
86
+
87
+ .
88
+ β”œβ”€ app.py
89
+ β”œβ”€ Dockerfile
90
+ β”œβ”€ requirements.txt
91
+ β”œβ”€ data/ # generated CSVs
92
+ └─ README.md
93
+
94
+ ## License
95
+
96
+ - Code: MIT (or your choice)
97
+ - Dataset: CC BY 4.0