Fixed vault-analysis-json chopping data

Author: miohtama

docs/source/tutorials/erc-4626-single-vault.ipynb

@@ -3993,7 +3993,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "id": "9225d6651c0a6852",
    "metadata": {
     "ExecuteTime": {
@@ -4011,7 +4011,6 @@
     }
    ],
    "source": [
-    "import pickle\n",
     "from pathlib import Path\n",
     "\n",
     "from eth_defi.vault.vaultdb import VaultDatabase\n",

docs/source/tutorials/erc-4626-vaults-per-chain.ipynb

@@ -497,8 +497,11 @@ def process_vault_group(group):
             # We may have severeal division by zero if the share price starts at 0
             warnings.simplefilter("ignore", RuntimeWarning)
 
-            lifetime_start_date = start_date = group.index.min()
-            lifetime_end_date = end_date = group.index.max()
+            # 2) Ensure group index is monotonic and clean
+            group = group.loc[~group.index.isna()].sort_index(kind="stable")
+
+            lifetime_start_date = start_date = group.index[0]
+            lifetime_end_date = end_date = group.index[-1]
 
             lifetime_return = group.iloc[-1]["share_price"] / group.iloc[0]["share_price"] - 1
 

eth_defi/research/vault_metrics.py

@@ -192,28 +192,16 @@ def find_non_serializable_paths(obj, path=None, results=None):
 vault_db = VaultDatabase.read()
 cleaned_data_parquet_file = PARQUET_FILE
 prices_df = pd.read_parquet(cleaned_data_parquet_file)
-
 print(f"We have {len(vault_db):,} vaults in the database and {len(prices_df):,} price rows.")
 
-# --------------------------------------------------------------------
-# Step 3: Filter data for the last N months
-# --------------------------------------------------------------------
-last_sample_at = prices_df.index[-1]  # Latest timestamp
-three_months_ago = last_sample_at - pd.DateOffset(months=MONTHS)
-PERIOD = [three_months_ago, last_sample_at]
-
-mask = (prices_df.index >= PERIOD[0]) & (prices_df.index <= PERIOD[1])
-prices_df = prices_df[mask]
-print(f"✅ Trimmed to {len(prices_df):,} rows from {PERIOD[0]} to {PERIOD[1]}")
-
 # --------------------------------------------------------------------
 # Step 4: Examine per-chain data availability
 # --------------------------------------------------------------------
 chain_ids = sorted(prices_df["chain"].unique())
 for chain_id in chain_ids:
     chain_name = get_chain_name(chain_id)
     print(f"\n🔍 Examining chain {chain_name} ({chain_id})")
-    chain_prices_df = prices_df[(prices_df["chain"] == chain_id) & (prices_df.index >= PERIOD[0]) & (prices_df.index <= PERIOD[1])]
+    chain_prices_df = prices_df[(prices_df["chain"] == chain_id)]
     print(f"📈 Rows: {len(chain_prices_df):,} for chain {chain_name}")
     if not chain_prices_df.empty:
         print(chain_prices_df.head(1))